CA1215521A - Catenated phosphorus materials, their preparation and use, and semiconductor and other devices employing them - Google Patents

Abstract

ABSTRACT

High phosphorus polyphosphides, namely MPx, where M is an alkali metal (Li, Na, K, Rb, and Cs) or metals mimicking the bonding behavior of an alkali metal, and x = 7 to 15 or very much greater than 15 (new forms of phosphorus) are use-ful semiconductors in their crystalline, polycrystalline and amorphous forms (boules and films). MP15 appears to have the best properties and KP15 is the easier to synthesize. P
may include other pnictides as well as other trivalent atom-ic species. Resistance lowering may be accomplished by doping with Ni, Fe, Cr, and other metals having occupied d or f outer electronic levels; or by incorporation of As and other pnictides. Top contacts forming function devices doped with Ni and employing Ni as a back contact comprise Cu, Al, Mg, Ni, Au, Ag, and Ti. Photovoltaic, photoresis-tive, and photoluminescent devices are also disclosed. All semiconductor applications appear feasible.
These semiconductors belong to the class of polymer forming, trivalent atomic species forming homatomic, cova-lent bonds having a coordination number slightly less than 3. The predominant local order appears to be all parallel pentagonal tubes in all forms, including amorphous, except for the monoclinic and twisted fiber allotropes of phosphor-us .
Large crystal monoclinic phosphorus (a birefringent material) in two habits, a twisted fiber phosphorus allo-trope and a star shaped fibrous high phosphorus material are also disclosed.
Single and multiple source vapor transport, condensed phase, melt quench, flash evaporation, chemical vapor depo-sition, and molecular flow deposition may be employed in synthesizing these materials. Vapor transport may be em-ployed to purify phosphorus.

The materials may be employed as protective coatings, optical coatings, fire retardants, fillers and reinforcing fillers for plastics and glasses, antireflection coatings for infrared optics, infrared transmitting windows, and optical rotators.

Description

S~L

-2-- llO-OOlB

TECHNICAL FIELD

This invention relates to ca~enated phosphorus mater-ials, their preparation and u'se, and to semiconductor and d other devices employing them. These materials include high phosphorus polyphosphides(i.~., phosphides where the polymeric nature is maintained), alkali metal polyphosphides, monoclinic phosphorus and new forms of phosphorus. Vapor transport is employed in making the crystalline, polycrystalline and amorphouY phosphorus and `
polyphosphide materials in bulk, thick and thin films.
Flash evaporation and chemical vapor deposition are used to make thin films. A condensed phase technique is utilized in producing crystalline and polycrystalline polyphosphides.
Diffusion doping is employed to raise the conduc~ivity of these materials. Rectifying junctions are formed wi~h the materials by appropriate metal contacts. The film materials may be used as optical coatings. Powdered crystals and amorphous materials may be used as fire retardant fillers.
The crystalline materials, especially the fi~rous forms, may be employed as the high tensile components of reinforced ~la~tics~ ;~
'~' ~;;
.
~, ..

-3- 110--OOlB
~2~ 521 BACKGROUND ART

Durinc3 the past several decade.s, the use of semi-conductors has become ever increasingly widespread and important. Silicon based semiconductors, for ex~nple, have generally been successful in pxoviding a variPty of useful devices, such as p-n junction rectifiers (diodes~, transis-tors, silicon control rectiiEiers (SCR's~, photovoltaic cells, light sensitive diodes, and the like. ~owever, due to the high cost of producing crystalline silicon and the ~:
ever-increasing demand for semiconductors over a broadening range of applications, there has been a need to widen correspondingly the scope of available useful semiconductor materials.
Useful semiconductors of the present invention, have an ~ ' energy band gap in the range of about 1 to 3 eV (more specifically 1.4 to 2.2 eV); a photoconductive ratio greater ~ -~than 5, (more specifically between 100 and 10,000); a conductivity between about 10 5 and 10 12 (ohm-cm~ 1 ~
(more specifically conductivity in the range of 10 8 to ~ `
10 9 (ohm-cm~13 and chemical and physical stability under ~ ,'!;
ambient operating conditions. Accordingly, while many materials may be semiconducting in the sense they are not pure metals or pure insulators, only those semiconducting materials which meet these criteria may be considered to be useful semiconductors in the context of this invention.
Given the present need to develop alternative non-petroleum based energy sources, the potential commercial utility of a semiconductor increases dramatically when the semiconductor also exhibits an effective photovoltaic characteristic, that is, the ability to economically and efficiently convert solar energy into electrical potential.
From an economic standpoint~ amorphous semiconductors, particularly in the fonm of thin films, are more desirable `~:`
than single crystalline forms due to potential lower cost of production. Amorphous semiconductors also have better electrical qualities than polycrystalline ~orms of the same -material all used in many semiconductor devices.

~4- 110-n0ls The semic~nductor industry has continued its search for useful new semiconductor materials beyond crystalline silicon, and the like.
In the non-silicon crystalline area, single crystals of semiconducting compounds, including GaAs, GaP, and InP, are in commercial use.
Many other semiconductor materials have be~n utilized for specialized purposes. For example, CdS and selenium are utili~ed as the photoconductor in many xerographic machines.
In this application semiconductor device means a device including a semiconductor material whether the device employs electrical contacts, that is, is an electronic device, or whether it is a non-electronic device, such as the photoconductors employed in xerography, phosphorescent materials, the phosphorus in a cathode ray tube, or the like.
Although some of the known forms of phosphorus have been stated to have semiconducting properties, many are unstable, highly oxidizable and reactive, and no known form of phosphorus has been successfully employed as a useful semiconductor.
The group 3-5 materials such as gallium phosphide and indium phosphide are tetrahedrally bonded and thus, as will be pointed out below, are clearly distinguished from the compounds disclosed herein. Furthermore, their semiconduc-ting properties are not dominated by phosphorus-to-phosphorus bonding, i.e. the primary conduction paths are not the phosphorus-to-phosphorus bonds.
Others have disclosed hydrogenated phosphorus having a structure similar to black phosphorus and having semicon-ducting properties.
Considerable worX on high phosphorus polyphosphides has been done by a group headed by H~ G. von Schnering. The various reports from this group indicate that the highest phosphorus containing polyphosphide compound they have pro-duced is cxystalline MP15 (M = group la metal). These polyphosphides are produced by heating a mixture of metal and phosphorus in a sealed ampule. Von Schnering reports that ba~ed on their structure polyphosphides ~5~1 llO-OOlB

are classified as valence compounds in a classical sense, and that this means that these compounds are, or should be, insulators or semiconductors, i.e not metals.
~ lonoclinic phosphorus, also called Hittorf's phos-phorus, is prepared according to the prior art from white phosphorus and lead as follows: lg of white phosphorus and 30g of lead are heated slowly to a melt in a sealed tube to 630~C and held for a short time at that temperature. The solution is then cooled at the rate of 10 per day for 11 days to 520C, and cooled ra~pidly to room temperature thereafter. It is next electrolyzed in a solution of 2kg of lPad acetate in 8 liters of 6% acetic acid, and the phos-phorus is collected in a watch glass placed under the anode.
Nearly square tabular crystals, about 0.2 x 0.2 x 0~05mm, are obtained in this way.
The structure of this prior art monoclinic phosphorus has been determined by Thurn and Krebs. The crystals comprise two layers of pentagonal tubes of phosphorus with all of the tubes parallel, and then another pair of layers of all pentagonal tube phosphorus, the tubes in the second pair of layers all being parallel, but the tubes in the second pair of layers being perpendicular to the tubes in the first pair of layers. The space group of the crystals has been determined, as well as the bond angles and bond distances. See the summary of the prior art in the section "Phosphorus" from "The Structure of the Elements" by Jerry Donahue, published in 1974.
The electronic properties of H~ttorf's phosphorus crystals have not been reported. Because of their small size the electrical properties cannot be readily determined.
The preparation of high purity electronic grade phos-phorus according to the prior art is very complex and time consuming, thus electronic grade phosphorus is very expen-sive, The prior art also exhibits a need for stable phosphorus compounds for use as fire retardants~
Crystalline forms have additional utility as reinforcing additives in plastics, glasses and other materials.

~Z~ 2~l -6- 110-OOlB

DISCLOSURE OF THE INVENTION
We have discovered a family of alkali metal polyphosphide materials possessing useful semicon~uctor, optical, and mechanical properti~es.
Useful Semiconductor Properties By "polyphosphide" we m~ean a material dominated by multiple phosphorus-to-phosphorus bonds. By l'useful semiconductor" we mean not only that the conductivity is intermediate between insulators and metals, but also the demonstration of a host of useful properties:
- Stability - Resilient m~terial struc-ture - Bandgap in a useful range (typically 1 to 2.5eV) _ High inherent resistivity, but with ability to be doped - Gooa photoconductivity - Efficient luminescence - Ability to form a rectifying junction _ Ability to be formed at relatively low temperatures (for semiconductors) by processes amenable to scale-up - Ability to be formed as large area amorphous thin films _ Ability to be formed as ductile polymeric fibers The polyphosphides are a unique family of materials possessing all of -these features.
Preservation of Utility in Multiple Forms It is equally significant that the useful properties remain essentially constant over a wide range of chemical compositions and physical forms (crystalline and amorphous).
To our knowledge, polyphosphides are the only useful semicon~uctors in which desirable single crystal-like 2~ llo-OOlB

properties are preserved in the amorphous form. This is of major technological significance because the amorphous form is at the least more amendable, and often essential, for large scale applications, such as photovoltaic cells, large area displays, and electrostatic copiers~
But up to now, the problem with amorphous semiconduc-tors is that they do not form readily as a stable single phase material. And even when lhey are forced to, the amor-phous form loses some very desirable features of its crys-talline counterpart.
The dominant known semiconductor (silicon~ has a tetra-hedral coordination in its crystalline form. Any attempt to make it amorphous (to make amorphous Si~ is known to be ac-companied by a breaking of the tetrahedral bonds, leaving "dangling bonds" that destroy useful semiconducting proper-ties. Pure amorphous Si is useless: unstable and crumbly.
Attempts to satisfy the dangling bonds with Hydrogen or Fluorine have been only partially successful.

Central Role of Structure We believe that the preservation of useful properties among the multiple forms of the polyphosphides are a direct result of the structure of the materials which, in turn, is made possible by the unique properties of phosphorus, particularly its ability to orm polymers dominated by 3 phosphorus to phosphorus covalent bonds at the vast majority of phosphorus sites.
In the crystalline form, the polyphosphides of the type MP15 ~with M = Li, Na, K, Rb, Cs) have a structure formed by a phosphorus skeleton consisting of parallel tubes with pentagonal cross section. These phosphorus tubes are linked by P-M-P bridges shown in Figures 4, 5 and 6. The building block for this MP15 atomic framework can be viewed as P8 ~ 21 llO-OOlB

(formed by 2 P4 rigid units) and MP7 (formed by the associa-tion of MP3 and P4 rigid units)r ~ sing the building hlocks or clusters described above, Kosyakov in a review article ~Russian Chemical Review, 48(2), 1979) showed theoretically that these polyphosphide compounds could be treated as polymeric materials using their basic building blocks as monomers. Hence, in principle it is possible to construct a large number of atomic frameworks having the same phosphorus skeletons.
In our work, we have synthesized by various techniques described later, MP15 crystals and also compositions of the type [MP7]a [P83b with b much greater than a.
These novel phosphorus rich compounds originally observed as "fibers", "whiskers", or "ribbons" are referred to in this investigation as MPX with x much greater than 15. These low metal content materials are prepared by vapor transport as thick films (greater than 10 microns) of polyc~stalline fibers and large boules (greater than 1 cm ) of amorphous character. The polycrystalline fibers exhibit the same morphology as KP15 whiskers.
The structural framework of the first MPX (x much greater than 15) crystalline materials we discovered is dom-inated by a phosphorus skeleton similar to the phosphorus framework of the MP15 compounds.
We have found that the useful electrical and optical properties of these crystalline materials MP15 and MPX
~x much greater than 15~ are similar. The properties of these materials are therefore dominated by the multiple P-P
covalent bonds of the phosphorus skeletons with a coordination number somewhat less than 3. To our surprise we have also discovered that the useful electro-optical properties of these materials were essentially preserved for MP15 and MPX (x much greater than 15) crystalline materials and their amorphous counterparts, Unlike previously known materials, this is a one dimen-sion rigid structure and is resilient in the following g lZ~ 110-OOlB

sense. The polyphosphide crystal symmetry is very low (tri-clinic). We believe that in the transi-tion from the crystal to the amorphous form, the low symmetry material is capable of accommodating in a gradual way the increased structural disorder that characterizes the amorphous state.
There is no ripping apart of strong tetrahedral bonds (coordination number of 4) as in silicon because the phosphorus, with a much lower coordination number than silicon, can accept much greater structural disorder without the creation of dangling bonds. The polyphosphides are polymeric in nature. The result is a polymeric amorphous structure with no apparent X-ray diffraction peaks, one with longer-range local order than is achievable with conventional amorphous semiconductors. We believe that this gradual onset, in the structural sense, of amorphicity is the reason for the preservation of the desirable crystal properties in the amorphous polyphosphides.
Distinguishable from Known, Useful Semiconductors The composition and structure of the family of polyphosphides clearly distinguishes them from all known, useful semiconductors:
Group 4a (Crystal Si, amorphous Si:H, etc.) 3a-5a (III-V) (GaAs, GaP, InP, etc.) 2b-6a (II-VI) (CdS, CdTe, HgCdTe, etc.) Chalcogenides (As2Se3) lb-3a-6a (CuInSe2) horus The alkali polyphosphides (MPX, M = Li, Na, K, Rb, Cs; where x = 15 and much greater than 15~ are phosphorus rich. In cases of "high x" material they are almost all phosphorus. Nonetheless, their structure (parallel pentagonal tubes) and their properties (stability, bandgap, 5~
~10- llO-OOlB

conductivity, photoconductivity) clearly distinguish them from all known phosphorus materials (black, white/yellow, red, and violet/ Hittorf). The structural relationships among these various forms are discussed below.
Our work has done much to clarify this aspect of phos-phorus itself. The nomenclature in this area has been some-what confusing. The following sllmmarizes our current usage.
1. Amorphous P or Re~l P
Amorphous red phosphorus is a generic term for all non-crystalline forms nf red phosphorus, usually prepared by thermal treatment of white phosphorus.
2. Violet P
This microcrystalline form of red phosphorus is prepared from charges o~ pure P, eith~r white or amorphous red, by extended thermal treatment.
3. Hittorf's P
Crystalline form of red phosphorus struc-turally identical to Violet P. Hittorf' 5 P
is prepared in the presence of a large excess of lead. Despite this, the terms "Hittorf's pn and "Violet P" have often been used inter-changeably. The crystal structure consists of double layers of parallel pentagonal tubes, with adjacent aouble layers perpendic-ular to each other in a monoclinic cell.
Hittor~'s P crystals are somewhat larger (approximately 100 microns) than violet P
microcrystals.

4. Large Crystal Monoclinic Phosphorus Even larger crystals (several mm), essential-ly isostructural with the above two, are described herein. These novel crystals are prepared by Vapor Transport (~T) treatments ~ Z ~ 110-OOlB

of alkali-phosphorus charges. The inclusion of the alkali is apparently essential for formation of the large crystals. Analysis confirms the presence of alkali (500 to 2000 ppm) in these large crystals of phosphorus.

5. Twisted Fiber Phosphorus A crystalline form of phosphorus described herein prepared by VT treatments of amorphous P charges.
Believed to be nearly-isostructural with polycrystalline MPX "ribbons".
ROLE OF THE METAL: WHY PHOSPHORUS IS ~OT GOOD ENOUGH
The many allotropic forms of elemental P are evidence for the variety and complexity of the bonds and structures that are accessible with phosphorus. We lack a detailed, comprehensive model of ~xactly how the alkali metal works, but have developed a ~arge body of data showing that the metal stabilizes phosphorus so that a single unique struc--12- ~ Z l llO-OOlB

ture may be selected from the ensemble of potentially available structuresr Without at least some alkali metal, the following undesired phenomena occur:
A. The phosphorus is unstable (e.g., White P).
B. To the extent that a known single phase is acces-sible to the P, it can only do so at high tempera-tures and of a size limited to microcrystals (e.g., Violet P), or C. At high pressures (e.g., black P).
D. Without an alkali metal in the charge, the MPX
type of structure is not formed by vapor trans-port. Rather, the twisted phosphorus fiber form we have discovered is obtained. This crystalline phase is metastable and the structure is not well defined as shown by our X-ray, Raman and photolum-inescence data.
The presence of alkali metal in vapor transport favors the all-parallel un.wisted phase. It also, as discovered by us, favors large crystals of monoclinic P to form at a different temperature.
The dominant role of structure, not composition, as the determinant of properties i5 made clear by noting that KPX
tx much greater than 15) has properties (bandgap, photolum-inescence, Raman spectra) that are essentially those of KP15, but are somewhat different from those of monoclinic It i-s clear that even a little alkali metal can serve to select a stable phase. But will non-alkalis work? Krebs reported non-alkali polyphosphides, with tubular structures consisting of 2b-4a-P14 (2b = Zn, Cd, ~g and 4a = Sn, Pb).
~hy do these form~
A speculative hypothesis is that these materials form in the tubular structure because the Group 4a element i5 amphoteric and car occupy a P site in lieu of P.
One can c:ompute an effective electron affinity of the P15 framework based on the ioni~ation ~nergies of the alkali metals, all of which are less than or equal to 5.1 3 ~ 2~ 110-OOlB

eV. One can, in turn, calculate effective ionization potentials for other possible compositions such as the 2b-4a-P14 compounds. All of Krebs' materials noted above have "effective ionization" less than or equal to ~.B eV.
USEFUL PRopERTIEs Our major initial discovery was that the KP15 whiskers (single crystal) wexe stable semiconductors, with an energy bandgap corresponding to red light (1.8eV) and exhibiting efficient photoconductivity and photoluminescence. These are the hallmarks of a semiconductor with potential applications in electronics and optics. Whiskers of the other alkali materials also have these properties (M = Li, Na, Rb, Cs).
To realize their potential, the materials had to be prepared in a size and form suitable for fabricating devices and for testing. We recognized tha~ the crystal habit, however, is not conducive to growth of large, single crystals that are frea of crystallographic "twinning".
Large, twin-free single crystals are the basis of nearly all semiconductor device technology today. Polycrystalline materials are less desirable because even if the individual grains are large, the presence of grain boundaries serves to destroy some desirable properties due to the physical and chemical discontinuities that are associated with such boundaries. Hence, our attention turned to the amorphous forms we had discovered.
Useful amorphous semiconductors, whether used as a junction device such as a photovoltaic cell, or as a coating such as in an electrostatic copier, have been generally made as thin films for extrinsic reasons (cost, manufacturing ease, and application need) and intrinsic reasons (material problems in the bulk amorphous state).
We have discovered that KP15 can be made as a stable amorphous thin film (by Vapor Transport). (This cannot be done with silicon: amorphous Si is not stable, while single crystal Si is.) Stable, bulk, and thin film amorphous KPX (x much greater than 15) can also be made by vapor transport.

-14- ~ llO-OOlB

There is evidence that these polyphosphides are unusual in yet another way. The useful properties of these materials MP15 and MPX (x much greater than 15) are similar in their crystalline forms and their amorphous counterparts as shown in Tables XVI and XVII below.
Applications utilizing amorphous thin film KP15 requiring no junctions can be readily envisioned (e.g., electrostatic copying). In factr the high inherent resistivity ~approximately 10 to 10 ohm-cm) is an advantage for such junctionless system applications.
Electronic and opto-electronic devices all require ~hat some junction be formed in or with the material.
This requires lowering the resistivity of the material by doping.
We have discovered that Ni diffused into KP15 serves the purpose of reducing the resistivity of the material by several orders of magnitude. Surface analysis has demonstrated that Ni diffusion from the solid state (KP15 deposited onto a layer of ~i) follows a normal diffusion pat-tern during the growth process of the film.
Device configurations with Ni as a back contact and diffuser; and other metals, such as Cu, Al, Mg, Ni, Au, Ag and Ti as top contacts, lead to junction formation.
Junction Current-Voltage (I-V) characteristics have been measured with these top contacts. Junction Capacitance-Voltage (CV) characteristics have been measured with Al and Au top contacts. The data indicates double junction formation with a high resistance layer near the top contact.
The high resistance layer is an undoped portion of the KR15 film which results from the present doping procedure.
A small photovoltaic effect (micro amp current under a short circuit condition) has been observed.
SY~THESIS OF POLYP~OSPHIDES
-Below are described the methods we have discovered that will produce polyphosphides of varying composition and morphology.

-15- llO-OOlB

A. Condensed Phase (CP~ S~nthesis This refers to the process of isothermal heat-up, soak (heating at set temperature), and cool down of starting charge carried out in a container of minimum volume. There is no vapor transport. Crystalline and bulk polycrystalline MP15 are produced.
B. Single Source Vapor Transport Synthesis (lS-VT) A starting reactant charge is located in one area of an evacuated tube which is heated to a temperature, Tc, which is greater than Td, where Td represents the temperature(s) of other area(s) of the tube where materials deposit from the vapor. Crystalline MP15; crystalline, polycrystalline (bulk, and thin films) and amorphous bulk high x, MPX; monoclinic phosphorus; star shaped fiber; and twisted fiber phosphorus are produced.
C. Two Source Vapor Transport Synthesis (2S-VT) Source reactant charges loaded in an evacuated chamber are separated physically by distance with a deposition zone between them. The two sources are heated to temperatures greater than the deposition zone (in order to get amorphous material, at least; see below). The deposition zone need not be the coldest one in the system, but a colder area should not be able to trap more than one component. 2S-VT was the first method used to make thin film amorphous KP15. Polycrystalline and amorphous thin films of MP15 and polycrystalline thin films and bulk amorphous high x, MPX are produced.
D. Melt Quench A charge is heated in a sealed evacuated tube (isothermally, if possible) to temperatures greater than the "melting point" as determined by endotherms observed in DTA
experiments, and held there for some period of time. The tube is the;n removed from the furnace and rapidly cooled.
CsP7 glass has been produced~
E. F1iash Evaporation A charge in powder form is fed in small amounts, under a slight Argon flow, into an RF-heated susceptor, which -16~ llO-OOls is maintained at temperatures greater than about 800C.
Inside the susceptor, the material is put thro~!gh a tortuous path where it is, in theory, forced to contact hot surfaces. This is intended tc rapidly and com-pletely vaporiæe the charge such that the compositionof the resultant vapor stream is the same as that o~
powder being injected. Th~_ vapor stream is directed into an evacuated chamber where it strikes cooler surfaces, resulting in condensed-product materials.
Amorphous films have ]been produced.
F. Chemical Vapor Deposition (CVD) In general, this refers to production of material by mixing two (or more) vaporized components which must undergo some chemical reaction to give products. As practiced by us, K and P4 are independently metered into furnaces where they are rapidly vaporized and ~ carried downstream by the Argon flow to a cooler reaction chamber where the combined streams yield condensed product materials.
The significance of CVD lies in that of these methods, it is the mos~ amenable to scale-up and to doping in situ~ i.e., simultaneous synthesis and doping of material. Amorphous thin films of KP15 have been produced.
G. Molecular Flow Deposition ~MFD) This is a multi-source vapor transport technique that draws on 2S-VT and Molecular Beam Epitaxy (MBE).
Independently heated sources are used and the vaporized species are allowed to reach the su~strate ~also independently heated) at a controlled rate not achievable with 2S-VT. The deposition takes place in an evacuated chamber with in situ monitoring of the deposition (also not available with 2S-VT). The chamber may be sealed or continuously evacuated to control pressure.

-17- llO-OOlB
5~
KP15 Materials A large variety of polyphosphide materîals of different physical forms and compositions were initially prepared during our investigations~
~ owever, for potential useful semiconductor applications, the emphasis of our work has changed from the preparation of single crystal materials to that of amorphous materials - either in bulk or large area thin films.
Among all the MP15 mat~_rials KP15 is a uni~u~
crystalline higher polyphosphidle (x the same as or greater than 7) compound which exists for the K-P system. (In contrast, the other alkali metals can form compounds with x = 7 or x = 11, such as CsP7, NaP7, RbP~
RPll and KP7 do not form as compounds. For this reason, the K-P system is easier to control than the other alkali-metal-P systems, where multiple compounds can form.
In addition, from the results of our experimental work, it is apparent that whenever K + P are vaporized, by whatever means, and brought in the proper ratio (lP]/lxJ the same as or greater than 15) to a zone whose temperature is in the proper window, amorphous ~P15 will form. By this window we mean the tempexature must be low enough to prevent crystallization of KP15 and high enough that KPX, where x is much greater than 15, is not deposited.
Based on this tenet, all synthesis methods can be seen to operate on the same general principle. Each method simply used different means to achieve control of source vaporization or control of deposition. The two source systems (2S-VT, CVD and MFD) are particularly useful as the important variables can be independently controlled.
Based on the above considerations, KP15 amorphous in thin films has been selected by us as our leading composition for the development of useful semiconductor materials.

-18~ S Z ~ llO-OOlB

SU~ARY

In a general inquiry into the nature of polyphosphides, potassium polyphosphide whiskers of about 1 cm in length were produced by single source vapor transport. In investi-gating the properties or this material it was determined by x-ray diffraction of a single c:rystal that the crystals were KP15. It was also discovered that these crystals were semi-conductors. When measuring an emission at 4K under argon ~
laser illumination, photoluminescence was observed having an `;
energy of 1.8 eV, thus indicating that the material possibly had a band gap within this energy range.
Later, in order to determine the conductivity of thesewhiskers, leads were attached with silver paint. In order to see if the leads were actually attached to a very small ~`
crystal, it was placed under a microscope while the conduc~
tivity was measured. Surprisingly the conductivity changed dramatically when the crystal was moved in the microscope, ~-changing the illumination. A photoconductivity ratio of 100 was measured with the unilluminated conductivity of the whisker being about 10 8(ohm-cm)~l To establish whether the whiskers had a band gap, measurements were then made of the wavelength aependence of the photoconductivity, the wavelength dependence of the optical absorption and th~
temperature dependence of the conductivity of the whisker.
These measurements, together with the photoluminescence -measurement at 4K, established that the whisXers had a band gap of approximately 1.8 eV. Thus it was established tha~
XP15 crystalline whiskers were potentially useful semiconductors.
An amorphous film was formed on the inside of the quartz tube during the vapor tran5port production of the KP whiskers. This amorphous film was also found to have -a band gap on the order of 1.8 eV and a photoconducti~ity ~ ' ratio on the order of about 100. Like the whiskers, the amorphous film had an electrical conductivity of approximately 10 8(ohm-cm) 1. Thus it was established that it also was a potentially useful semiconductor.

~Zl~ llO-OOlB
The problem then presented to the inventors was whether KP15 could be producecl as large crystals, such as silicon, used in semiconductor production; whether polycrystalline or amorphous films of KP15 could be reproducibly made and utilizecl for semiconductor production;
and the full characterization of the materials produced by the vapor transport experiment and any analogous materials which might have the same useful properties.
After many vapor transport experiments the inventors were astonished to find that the polycrystalline and amorphous materials that were produced by vapor transport where a single source of a mixture of potassium and phosphorus is heated and material condensed at the other end of a closed tube, were not KP15 but when measured by wet analysis were KPX where x seemed to range from about 200 to about 10,000.
The inventors have since made the amazing discovery that the affinity of phosphoru~ for potassium, or any alkali metal for that matter, in single source vapor transport causes initial deposition of as the most stable polyphosphide. If there is an excess of phosphorus, then a new form of phosphorus will be deposited. (MPX where x is much greater than 15) This new form of phosphorus has the same electronic qualities as KP15 and is a useful semiconductor.
During the course of their investigations the inventors, in an effort to form thin films of polycrystalline and amorphous KP15 and other alkali metal analogs which could not be formed by single source vapor transport, conceived of a two source (separated source) vapor transport method in which the alkali metal and the phosphorus are spaced apart and separately heated. By controlling the temperature of a separate intermediate deposition zone, thin films of MP15 where M is an alkali metal, have been made in polycrystalline and amorphous forms. This technique has also led to the production of thin films of polycrystalline and thick films of amorphous phosphorus material of the new form, and other materials presumably polymer-like having the formula MPX, where M is an alkali metal and x is much greater than 15.

_~o~ ~ Z ~ 52 i llO-OOlB

We have also used Flash Evaporation, Chemical Vapor DepositiOn, and propose to use Molecular ~low Deposition methods for synthesizing these materials.
We use MPX as the formula for ~1] pol~phosphides. As will be pointed out below, for useful semiconductors, x may range from 7 to infinity. Known alkali polyphosphides have the formula MP7, MPll, and MP15. We have discovered that presumably polymer fo~ms exist having the formula MPX
where x is much greater than 15.
Also during these invest:igations single source vapor transport has been improved over the pxior art by control-ling the deposition temperature to be constant over a large area, so that large area thick films and boules of polycrys- ~
talline and amorphous MPX where x is much greater than 15 -have been formed.
Large quantities of crystalline and polycrystalline MP15, where M is an alkali metal, have been made by isothermally heating toget~er stoichiometric proportions of an alkali metal and phosphorus. This condensed phase method produces excellent NPX where x ranges from 7 to 15 for use in single source vapor transport. The condensed phase method itself is facilitated by the prior mixing and grind~
ing together an alXali metal and phosphorus in a ball mill which is preferably heated to a temperature in the neighbor- ~;
hood of 100C. This ball m~lling surprisingly produces ~ ~
relatively stable powders. ~:;
All of the parallel tube polyphosphides have a band gap of approximately 1.8 eV, photoconductivity ratios much -greater than 5, (measured ratios having a range from 100 to 10,000), and low conductivity in the order of 10 8 to 10 9(ohm-cm) L.
Since we have discovered that the amorphous forms of these materials, i.e. alkali polyphosphides MPX where x is greater than 6 formed in the pre.sence of an alkali metal have substantially the same semiconductive properties~ we conclude that the local order of the amorphous materials is ~-thesc~e~i~é~c~l parallel pentagonal tubes substantially throughout their extent.

5Zl -21- llO-OOlB

In all the polyphosphides~ the 3 phosphorus to-phos-phorus (homatomic) covalent bonds at the majority of phosphorus sites dominate any other bonds present to provide the conduction paths and they all have semiconductor properties.
The covalent bonds of the phosphorus atoms,all of which are used in the catenation providing the dominant conduction paths and the parallel local order in these materials,pro-vide the good semiconducting properties. The phosphorus atoms are trivalent and the catenations form spiral~ or tubes having channel-like cross sec~ions. The alkali metal -atoms, when present, join the catenations together. Atomic species other than phosphorus, particularly trivalent spe~
cies capable of forming 3 covalent homatomic bonds, should also form semiconductors.
Thus we have invented new forms of phosphorus and methods of making the s&me, solid films of amorphous and polycrystalline MPX and methods and apparatus for making the same~ methods and apparatus for makin~ metal polyphosphides by multiple temperature sin~le source techniques, methods and apparatus for making high phosphorus `~ ~;t polyphosphides by multiple ~eparated souxce techniques, methods and apparatus for making MP15 by condensed phase techniques in polycxystallin forms, semiconductor devices comprising polyphosphide groups of seven or mor2 p~osphorus atoms covalently bonded together in pentagonal tubes having a band gap greater than 1 eV and photoconductivity ratios of 100 to 10,000, semiconductor devices compr;sin~ MæX whera M is an alkali metal and x is greater than 6, and materials having a band gap greater than 1 eV and photoconductivity ratios of 100 to 10,000, semiconductor devices formed o~ a high proportion of catenated covalently bonded trivalent atoms, preferably E)hosphorus/where ~he catenated atoms are joined together in multiple covalent bonds,the local order of which comprises layers of c~tenated a~oms which are parallel in each :Layer and the layers are parallel to each other, the catenations preferably bein~ penta~onal tubes~
semiconductor devices comprising an alkali metal and said '':`
.
.

-22- llO-OOlB

catenated structures wherein the number of cor.secutive covalent catenated bonds is sufficiently greater than the number of non-catenated bonds to render such material semiconducting, semiconductor devices formed of compounds comprising at least two catenated units, each unit having a skeleton of at least 7 covalently bonded catenated atoms, preferably phosphorus,~ and having alkali metal atoms conductively bridging the skeleton of one unit to another, junction devices, methods of forminy such semiconductor devices, methods of doping such semiconductor devices, methods of conducting electrical current and generating electrical potential utilizing such devices.
We have therefore discovered a whole class of materials to be useful semiconductors, some members of the class having been first produced or properly characterized by us, and others of which have been produced in the prior art with ~heir useful semiconductor properties being unknown until our discoveries and inventions.
All of these materials have a band gap within the range of 1 to 3 eV, pref~rably within the range of 1.4 to 2.2 eV
and most preferably about 1.8 eV. Their photoconductivity ratios are greater than 5 and actually range between 100 and 10,000. Their conductivities are within the range of 10 5-10~12(hm-cm)-l being in the order of 10 8(ohm-cm3-1.
Those skilled in the art will readily understand that the alkali metal component M of polyphosphid~ or any appropriate trivalent "ide" capable of forming homatomic covalent bond50 and having the formula MYX ma~ comprise any number of alkali metals, (or combination of metals mimicking the bonding behavior of an alkali metal) in any proportion, without changing the basic pentagonal tubular structure and thus without signi~icantly affecting the electronic semiconductor properties of the materialD
We have further discovered and invented methods of doping the materials of the invention utilizing doping with iron, chro~ium and nic~el, to increase the conductivity.
Junctions have been prepared using Al, Au, Cu, Mg, Ni, Ag, Ti, wet silver paint, and point pressure contacts.

-23- llO-OOlB

The incorporation of arsenic into the polyphosphides (all parallel tubes) has also been demonstrated to increase conduct ivity .
These doping methods are also part of our invention and discovery.
The semiconductor materials; and device~ of the present invention have a wide variety of uses. These include photoconductors such as in photocopying equipment; light emitting diodes; transistors, diodes and integrated cir- ;
cuits; photo~oltaic applications; metal oxide semiconduc-tors; light detection applications; phosphors subjected to ~::
photon or electron excitation; and any other appropriate semiconductox application. - :~
In the course of our work we have also produced for the ~
first time large cFystals of monoclinic phosphorus~ These ~ -:
crystals are obtained from vapor transport technique using an :
MP15 chargeora mixture of M ~nd P (M/P) in varying ratios.
Surprisingly, these lar~e crystals of monoclinic phosphorus contain a significant amoun of alkali metals (500 to ~000 ppm have been observed). Under the same conditions, these :
crystals cannot be grown without the presence of alkali metals in the charge. ~ :
Two different crystal habits have been observed for these large crystals of phosphorusr 2 One crystal habit was identified as truncated pyramidal ~' shape crystals as shown in Pigur~ 3g. These crystals are -~.
hard to cleave. The other form is a platelet-lik~ cry~tal and is cleavable as shown in Figure 40. ~
The largest crystals we have produced in the habit ~-:
shown in ~igure 39 are 4 x 3mm x 2mm high~ The laxge~t crystals we ha~e produced in the habit shown in Pigure 40 :. ~
are 4mm square anZ 2mm thick. ! - ' The cryst:als are metallic looking on reflection and ~;
~eep red in transmission. Chemical analysis indicates that they contain anywhexe from SQ0 to 2000 parts per million of alkali metal. Their powder X-ray diffraction p~tterns, Raman spectra and differential thermal analy~is are all con~istent wit~h ~he prior art Hittorf's phosphorus.

' ..
'-I

-24~ Z 1 llO-OOlB

Photoluminescence of crystals grown in the presence of Cesium in Figure 41 and crystals grown in the presence of Ru~idium in Figure 42 show peaks at 4019 and 3981cm 1, which indicate a band gap of about 2.1 eV at room temperature for this monoclinic phosphorus.
The crystals may be utilized as a source of.phosphorus;
as optical rotators in the red and infra-red portio~ of the spectrum (Shey are hirefringent); as substrates for the growth of 3-5 materials such an Indiu~ Phosphide and Gallium Phosphide, They may be utilized in luminescent displays or as lasers.
We hava grown from the same charge and deposited at a slightly lower temperature the star shaped fibrous crystals shown in Figures 44 and 45.
We have also grown by vapor transport a crystal allotrope of phosphorus, the twisted fiber of phosphorus shown in Figure 46.
The polyphosphides may be used as fire retardants and strengthening fillers in plastics, glasses and other materials. The twisted tube and star shaped fibers ~hould be of particular value in strengthening composite materials because of their ability to mechanically interlock with the surroundin~ material, The platelets should be of particular value in thin sheet material where glass flakes are now employed The film materials of the invention may be utilized as coatings for thair chemical stability, fire retardant and optical properties.

~z~
-25- llO-OOlB

OBJECTS OF THE IMVENTION
.
It is therefo~e an object of the invention to provide z new class of useful semiconductor materials.
Other objects of the invention are ~o provide new meth-ods and apparatus for making polyphosphides.
Still other objects of the invention are to provide new forms of stable high phosphorus materials and methods and apparatus for making the same.
Further objects of the invention are ~o provide new fonms of phosphorus and methods and apparatus for making the same.
Still other objects of the invention are to provide dopants and methods of doping such materials, Yet other objects of the invention are to provide semi~
conductor devices employing the above.
Another object of the inventionisto provide large crys-tals of monoclinic phosphorus.
Still another object of the invention is to provide high purity phosphorus.
Still another object of the inventio~ is to provide new semiconductor materials.
Still another object of the invention is to provide a birefringent material for use in the red and infra-red por-tion of the spectrum.
Yet still another ob~ect of the invention is to provide methods for making materials of the above character.
A further object of the invention is to provide such m2thods which are more convenient than the prior art and less expensive.
Other objects of the i~ention are to provide coating materials, fillers, reinforcin~ materials, and fire retard-ants.
Other objects of the invention will in part be ob~ious and will in part appear herein~ftex, ~ he invention accordingly comprises one or more inven-tive steps and the relation of such steps with respect to ea~h of the others which will be exemplified in the methods -26- ~L2~ o-~ol~

and processes hereinafter describcd, compositions of matter possessing the characteristics, properties and the relation-ship of constituents and componer,ts which will be exempli-fied in the compositions hereinafter described, articles of manufacture possessing the features, properties, and the re lation of elements which will be exemplified in the articles hereinafter described and apparatus comprising the features of construction and arrangement of par~s which will be exem-plified in the apparatus hereinafter described. The scope of the invention is indicated in the claims.

-27- ~L llO-OOlB

BRIEF DESC~IPTION OF THE VRAWINGS
. . . ~
For a fuller understanding of the "ature and objects of the invention, reference shoulcl be had to the following detailed description, taken in connection with the accom-panying drawings, in which:
FIGURE 1 is a diagr~Nmatic view partly in cross section of single source vapor transport apparatus according to the invention;
FIGURE 2 is a diagrammatic view of a portion of the vapor transport apparatus of FIGURE 1, FIGURE 3 is a diagrammatic view of another s.ingle source vapor transport apparatus- according to the invention;
FIGURE 4 is a computer diagram from X-ray diffraction data of phosphorus atoms in MP15 where M is an alkali metal;
FIGU~E 5 is a computer diagram from X-ray diffraction data of a cross section of KP15 showing how the covalent bonding of the phosphorus atoms of FIGURE 4 forms a pentagonal tu~ular structure;
FIGURE 6 is a computer diagram from X-ray diffraction data in longitudinal section of KP15;
- FIGIJRES 7 and 8 are photomicrographs of KP15 crystal whiskers;
FIGURE 9 is a powder X-ray diffraction Eingerprint of crystalline KP15;
FIGURE 10 is a powder X-ray diffraction fingerprint of crystalline KPX where x is much greater than 15;
FIGURE 11 is a diagrammatic view of an experimental reaction tube for t.wo source vapor transport according to the invention;
FIGURE 12 is a plot of temperature versus length for the reaction tube of FIGURE 11;
FIGURE 13 is a diagram of the P to K ratio of the reaction products in the reaction tube of FIGURE 11;
FIGURE 14 is a ~chematic diagram of apparatus for a two source vapor transport according to the invention;

-28~ llO-OOlB

FIGURl, 15 is a diagram of onc of the elements of the apparatus illustrated in FIGURE 14;
FIGUR~ 16 is a diagranmatic view of another reaction tube for two source vapor t:ransp~rt according to the inven-tion;
FIGURE 17 is a diagrammatic view of a ball mill accord-ing to the invention;
FIGURES 18, 19 and 20 are scanning electron micrographs (SEM's) of a film of a new form of phosphorus MPX where x is much greater than 15;
FIGURE 21 is a photomicrograph of an etched amorphous surfacs of such high x MPX synthesized b~ single source vapor transport according to the invention;
FIGURE 22 is an photomicrograph of an etched amorphous surface of such high x MPX synthesi~ed by two source vapor transport according ~o the invention;
FIGURE 23 is an photomicrograph of the same surface as shown in FIGURE 22;
FIGURE 24 is an photomicrograph of an etched surface perpendicular to the surface shown in FIGURES 22 and 23;
FIGURE 25 is an SE~ photomicrograph of the upper surface of an amorphous thin film of KP15 svnthesized by two source vapor transport according to the invention;
FIGVRE 26 is a cross sectional view partly in diagram-matic form illustrating the formation of a junction according to the invention;
FIGURE 27 is an illustration of the oscilloscope screen in the experiment illustrated in FIGURE 26, and FIGURE 28 is a cross sectional view partly in diagram-matic ~orm illustrating the formation of a junction according to the invention;
FIGURE 29 is an illu5tration of the oscilloscope screen in the experiment illustrated in PIGURE 28;
~ IGURE 30 is a diagram of a photosensitive resistor according to the invention;
FIGURES 3:L, 32, and 33 are illustrations of oscillo-~cope screons showing junction activity in devices according to the inv~ntion; .

-29~ 21 llO-OOlB

FIGUR~S 34, 35 and 36 are plots of capacitance versus applied electrical potential in junction devices according to the invention;
FIGUR~ 37 is a plot of capacitance and resistance as a function of fre~uency of applied potential of devices according to the invention;
FIGURE 38 is a diagram of a preferred form of sealed ampoule utilized to form monoclinic phosphorus according to the invention;
FIGURE 39 is a photomicrograph of a crystal of mono-clinic phosphorus according to the invention;
FIGURE 40 is a photomicrograph of a crystal of mono~
clinic phosphorus according to ~he invention;
FIGURE 41 is a diagram of the photoluminescence response of a crystal of monoclinic phosphorus according to the invention;
FIGURE 42 is a diagram similar to FIGURE 6 of the photoluminescent response of a crystal of monoclinic phos-phorus according to the invention; and FIGURE 43 is a Raman spectrum of monoclinic phosphorus.
according to the invention.
FIGURE 44 and.45 are SEM photomicrographs of another new form of phosphorus according to the invention;
FIGURE 46 is an SEM photomicrograph of still another new form of phosphorus according to the invention;
FIGURE 47 is a side diagrammatic view of flash evapora-tion apparatus according to the invention;
FIGURE 48 is a cross sectional view taken along the line 48-48 of FIGURE 47;
FIGURE 49 is a cross sectional view taken along the line 49-49 of FIGURE 48; and, FIGURE 50 is a diagram of chemical vapor deposition apparatus, according to the invention.
The same reference numbers refer to the same elements throughout the several views of the drawings, , .

~2~52~

-30- 11 O -OOlB

BEST MODE FOR CARRYI~G OUT_THE INVENI~IoN

The high phosphoxus materials of ~:he il~ven~ion exempli-fied by the high phosphorus l?Olyphosphides ~PlS where M isan alkali metal, and the new forms of phosphorus formed, are all believed to have similar lo~al order, whether crystal-line~ polycrystalline or amorphous. We believe that in both crystalline and amorphous MP]5, this local order takes the form of elongated phosphorus tuhes having pentagonal cross: :-sections as shown in Figures 4, 5 and 6. All o the pentagonal tube5 are generally parallel on the local scale and in MP15 double layers of the pentayonal phosphorus tubes are connected to each other by interstitial alkali metal atoms. In the new forms of phosphorus o~ our inven-tion, many, i not most of the alkali metal atoms are missing. ~owever, it appears that one new form of phos-phorus formed in the presence of very small amount~ of alkali metal atoms grows from vapor deposition in the same form as MP15 One experiment to be discussed bPlow indi~
cates that at least one form of this is by growth of the new form of phosphorus on a layer of MP15. The MP15 may act as a template causing the phosphorus to organize in the same structure. All o the materials having these all parallel pentagonal phosphorus tubes have been ~ound by us to have a band gap between 1.4 and 2.2 eV and most o~ the order of 1.8eV. Photoconductivity ratios range from 100 to 10,000.
Thus it is indicated that all high phosphorus alkali metal polyphosphides from MP7 throu~h MP15 and more complex forms and ~ixed polymers of MP15 and the new form of phos-phorus dis¢overed by us (MPX wher2 x is much greater than lS), which all have the all paralle] pentagonal tu~e structure, if stable, will be useful semiconductor materials, barring the inclusion of elements that would act as traps, cause the formation of grain boundaries, or the lik~.
In all of these materials having 1:he all parallel pen~
tagonal tubular s~ructure, our investigatlons indicate ~.
.

~z~
-31- llO-OOlB

that the multipl~ continuous covalent phosphorus to-phosphorus bonds of the tubes being substantiall~ greater in nllmber than the number of other bo~ds will provide primary electrical conduction pa~hs for electrons and holes and thus p~ovide good semiconductor properties. It is further our opinion that the presence of alkali metals in he charge, even when resulting in trace amounts in the new forms of phosphorus we have discovared, promote growth of the materials in forms that maintai.n the same structural and electronic properties as KP15 or as monoclinic phosphorus, depending on deposition conditions.
The family of semiconductor members to which th~
subject invention is directed comprises high phosphorus polyphosphides ha~ing the formula MPX whexein M is a Group la alkali metal, and x is the atomic ratio of phosphorus-to-metal atoms, x being at least 7. Metallic elements of Group la most suitable are Li, Na, K, Rb, and Cs~ Although francium presumably is suitable, it is rare, has not been involved in any known synthesis of MPX and ~s radioactive.
High phosphorus polyphosphides where M includes Li, Na, K, Rb or Cs have been formed and tested by the inventors.
The polyphosphide compound5 of this invention as presently defined mu~t contain an alkali metal. Some of the new forms of phosphorus must be formed in the presence of minor amounts if-not unmeasurable amounts of alkali metal.
However, other metals may b2 presen~ in minor amounts as, for example, dopants or impurities.
KP15 and, as we later learned, a new ~orm of pho~phorus was first synthesized as follows.
Referring to Figure lt a two temperature ~one furnace 10 comprises an outer sleeve 12 preferably constructed of iron~ Outer sleeve 12 is wrapped in a thermally insulative coating 14 which can comprise an asbestos clothc ~he furnace was c:onstructed in the laboratory shop of the inventors.
We used a P~K atom ratio of about twelve ~123 as reacta~ts 36 in furnace 10. As one illustrative example, 5.5 g of red phosphorus and O.~ g of potassium were trans--32- i Zl ~ S Z ~ llO-OOlB

ferred und~r nitrogen to quartz tube 32. Prior to transfer, the phosph~rus was washed repeatedly with acetone, and air dried. However, this washing is considered optional, as is the solvent selected.
After being charged with reactants 36, tube 32 was evacuated to, for example, lO 4 Torr, sealed, and then placed in furnace lO. Tube 32 was mounted at a slight incline in the furnace. Power s~spplied to conductors 24 and 2~ was adjusted to establish a 1:emperature gradient of, for example, 650DC to 300C from heat zone 28 to heat zone 30.
With the above described inclination of furnace ~0, react-ants 36 were assured of being located in the hotter temper-ature heat zone 28.
After maintaining furnace lO at these conditions for a sufficient period of time, for example approximately 42 hours, power to conductors 24 and 26 was terminated and tube 32 was allowed to cool. Upon reaching ~mbient temperature, tube 32 was cut open under a nitrogen atmosphere and the contents of tube 32 were removed. The contents were washed with CS2 to remove pyrophoric materials, leaving approxi-mately 2.0g of stable product. This resulted in a yield of approximately 33 percent.
Using this form of synthesisS various phases of resul-tant product occur at well defined positions within ~ube 32 as illustrated in Figure 2. A dark gray-blac~ residue 40 coupled with a yellow-brown film 42 is typically produced at the extreme end of hot zone 30, where reactants 36 are initially located. Moving in a direction of decreasing temperature along tube 32, there is next found black to purple film deposit5 42 which are a polycrystalline material. Next to film deposits 42 is an abrupt dark ring of massed crystallites 44 and immediately adjacent crystal-lites ~4 is a clear ~one wherein whiskers 46 are grown. A
highly reflective coating or film aepo~it 48 is found on the lower portion of tube 32 in the beginning of cold zone 28.
Above film deposit 48 a deep red film deposit 50 occasion-ally occurs ~epending on ~he temperature maintained in the zone, The deposits 48 and 50 can be polycrystalline, ~Z~
-33- llO-OOlB

amorphous or a mixture of polycrystalline and amorphous material d~pending on the reactants and temperature. At the extreme end of cold zone 28 ~s a mass or film deposit 52 which is amorphous material.
Since ~here is a continuous temperature gradient from the hot zone to the cold zone of the reaction tube shown in Figures 1 and 2, the nature of the materials deposited actually varies conti~uously from high quality crystalline whiskers to polycrystalline to amorphous. In order to ma~ipulate the reactio~ and attempt to deposit large areas of uniform layers of material, a three zone furnace was :~
constructed and is illustrated in Figure 3. As herein embodied, the three zone furnace.54 is essentially identical to furnace 10 illustrated in Figure 1, in that furnace 54.
comprises an outer iron sleeve 56, a tube 60, and a reaction tube 5B. For purposes of simplicity, asbestos wrappings of outer sl~eve 56 and tube 58 have been omitted from ~igure 3.
Furnace 54 is primarily di5tinguishable from furnace 10 in ~ ' that ~ube 58 is much longer in comparison to tube 32, and ls .
preferably on the oxder of 48 cm in length. In addition, ~-furnace 54 has associa~ed wi~h it three distinct heat zones, `.
62, 64 and 66 which ar2 individually controllable to create a more definitive heat gradient along tube 60. Tube 60 may be supported by asbest~s blocks 68 and 70 in a manner so as to provide for an inclination of tube 60 an~ reaction tube 58 toward heat z~ne 62, in ord~r to keep reactants 36 in -;
proper position.
Very good quality preparation of KP15 whiskars were obtained usi~g temperature set points of 550, 475, and 400 degrees centigrade in heat zones 62, 64 and 66 respectively. . . :
It was also found that bulky deposits ~enerated in furnace 10, wh~n loacled into inner slee~e 60 of furnace 54 and reheated in the above identified temperature gradient, would sublime to form film deposits like tho5e o~ fi~ms 48-52 :~:
illustrated in Figure 2, but only when a high zone tPmpera- :
ture of ~t least 400-475C was used.
Unit cell struct~ral information on XP15 ~rystals produced in accordance with the method described above was .

-34~ llO-OOlB

obtained by single crys~al X-ray diffraction data, and collected with an automated diffractometer. A fibrous single crystal of 100 microns diameter was selected and mounted on a glass fiber. The structuxe was determined by direct methods usin~ a total o~E 2,544 independent reflec-tions. All the atoms were located by an electron map and differential Fourier synthesis.
Typical needle-like crystals were examined by high magnification and scanning electronic microscopy (SE~). The resultant SEM photographs of the cross section o the needles show that the needles are apparently composed of dense fibrils rather than hollow tubes. Marked twinning of the whisker crystals is also discernible on the microphoto-graphs of KP15 whiskers in Figures 7 and 8. The diam~ter of the primary fibrils of the whisker-type crystals is esti-mated to ~e approximately 0.1-0.2 microns. J.arger ibrils seem to have a ine structure consistin~ of parallel lamellae of approximately 50~ angstroms thickness.
From the initial crystal data ~efinement study 9 the stoichiometry of the studied potassium phosphide compound appears to be XP15.
The phosphorus atomic framework of the compound is formed of identical unit tubes with a pentagonal cross section. The tubes are unidimensional along the needle axis direction. The phosphorus tubes are parallel to one an-other. In the simplest descrip~ion, double layers of separated phosphorus tubes are ~onnected by a layer of potassium atoms. As judged by the inter atomic distances, the K atoms are at least partially ionicall~ bonded to P
atoms. A cross sectional view of a whisker is presented in Figure 5.
More specifically, each potassium site is associated with a rigid unit of 15 consecutive phosphorus atoms having a structure as illustrated in Figure 4. In this rigid unit all the phosphorus atoms but one are bound to three other phosphorus atoms. The other phosphorus a oms are chained, with the missing bonds linked to a potassium atom as showr~
in Figure 5. ~hus, the potassium atom appears to link -35~ llO-OOlB

tubular phosphorus units through a missing P-P bridge. In the investigated structure, potassium has phosphorus atoms as nearest neighbors at distances of 3.6A, 2~99A ~nd 2.76A, respectively. The P-P distances vary from 2.13A to 2.58A.
The bond angles at the phosphorus chains vary between 87 to 113~ and average 102~ ~
Arsenic forms a layered structure having an average bonding angle of 98 and this is not known to be a useful semiconductor. Black phosphorus has a similar stxucture and an average bonding angle of 96~. Trivalent atoms which can form their three bonds within the range of 87 to 113 with the average above 98 may form the same catenated structure as MPX. If the bonds are covalent the material can be expected to have the same electronic properties as MPX.
Table I gives the crystal lattice parameters and atomic positions we found for crystalline KP

TA~T.F: I

Crystal Lattice Parame~ers For KP15 ... . . ..
Triclinic system Unit cell parameters .

a = 9.087 A~ (~ 0~15) A
b = 11.912 A (+ 0.10) A
c = 7.172 A (+ 0.15~ A
= 101.4 ~+ 0.13~
= 107.9 (+ 0,2~
~r_ ~9.3 ~+ 0.1) The unit cell is primitive with one molecule pex unit cell and a volume of 723~3 Cubic Angstroms , --36-- ~23L~$21 110--001B

Space group Pl The high st attainable symmetry in 1:he above structural configu~atiOn is a centrosymmetric Pl space group with the stoichiometry given by KPl5.
The corresponding X-ray powder diffraction data for XPl5 polycrystalline material with copper illumination is shown in Pigure 9. This shows the d spacing with the correspondin~ X-ray intensities.
Similar X-ray powder diffraction data have been obser-ved for whiskers and polycrystalline MPl5 materials with M= Li, Na, K, ~b and Cs.
In all these isostructural compounds, the structural framework can be viewed as formed of parallel pentagonal phosphorus tubes. These tubes are linked by a P-M-P bridge.
The rigid units for this type of structure are P~ and MP3. The building block for the atomic framework can be viewed as [P4 - MP3] or [MP7].
Therefore:
lMP7l ~ 2[P4] ~ [P4 M 7 which represents the basic structure MPl~.
Of course, one of the building blocks in such compounds may be present in much larger quantities than the other. In the case of MPX, for example, there may exist building blocks of lMP7] and lP8], which are present in a ratio of a to b, respectively. In such a case MPX could be ex-pressed in the form [MP7]a[P8~b, wherein mathematically x = 17a~8b)/(a).
It is also possible for a compound to have b much greater than a and have the same basic structural framework.
This type of polymer like tubular structure will result in "fibers" or whiskers of the type MPX with x much greater than 15.
Whiskers and polycrystalline "fibers" of the type MPX
with x greater than lOOO (M=Li, Na, R, Rb, Cs) have been observed to crystallize at low temperature (abou~ 400~C) using the vapor transport technique~ The X-ray powder diffraction data of these materi,als are substantially the same. Data for KP~ where x is much gxeater than 15 under copper illumination is shown in Figure lO.

-37- 12i~ llO-OOlB

We can compare the structure described above to other structures based on pentagonal cross section phosphorus tubes. The KP15 compound is isostructural to LiP15' NaP15' RbP15' CSP15 The other alkali metals appear to play the same role as K.
From structural data we concluded that numerous compounds can be formed which will be based on pentagonal cross section tubular building blocks. We also found that in phosphorus materials, at least partially, the phosphorus atoms can be replaced by other pnictides, such as As, Bi or Sb. Substitution under 50 atom percent is possible, without adversely affecting the basic stxucture of the high phosphorus polyphosphides.
In Table II are shown the various MPX compounds synthesized that we have found the same structure as crystalline KP15 as shown by XRD powder diffraction fingerprint analysis.
Table II
Rigid units MP3 and P4 Building blocks [P4-MP3~ or ~MP7] and [P8]

Basic structure [ 4 MP7-P4] or [MP ]

25 M: Li, Na, K, Rb, Cs Compounds Isostructural With Crystalline KP15 M M' P'y P
with: O <- x ~ 1 y < 7.5 M and M' from Group la P' from Group 5a (As, Bi, Sb) , S;~l -38- llO-OOlB

Initially the inventors found, as previously stated, that the crystalline whiskers produced in the apparatus of Fisures 1, 2 and 3 were MP15~ However, analysis of the polycrystalline and amorphous materials, although indicating that these materials had the sa~ne semiconducting propert.ies as the MP15 whiskers, had widely variable stoichiometric P200 to MP]o,oOO, and Surprisingly no manipulation of the temperatures in the three zone furnace illustrated in Figure 3 would- produce amorphous form~ of MP15. It was therefore necessary to greatly refine the methods of producing these materials and to invent a new two source vapor transport apparatus in order to successfully produce polyerystalline and amorphous MP15 materials. The very high x materials which are now thought to be a new form of phosphorus, have also been prepared by this method by initially depositing MP15 and thereater cutting off the source of alkali metal so that only phosphorus vapor is present for deposition of p~osphorus. Additionally~a . :~' condensed phase process has been extensively investigated using molar charges of ~Px materials where x varies from 7 to 15. In this method the stoichiometxic mixtures are heated isothermally to reaction and then cooled. We have produced a wide variety of MPX
materials i~ this manner which are crystalline or polycrystalline powders.
There follows a detailed description of the methods we hav~ amployed to synthesize high phosphorus materials and how we have measured the electro-optical characteristics and dcmonstrated that they are useful semiconductors.

'' ~' 1 -39- 12~ 1 llO-OOlB

Preparation of Stable High Phosphorus ~aterials by the Vapor Transport Techni~ue from a 5ingle Source Introduction -.
The technique of applying sufficient energy to a system to create vapor species which yield products on condensation or deposition, at appropriate temperatures,is oalled "vapor transport". For the followins discussion, where the source materials are held in close contac~ and heated together at about the s~me temperature, the ~urther descriptio~ as a "single source" technique is applicable.
The methodology described by Von Schnering was essen-tially a single-source vapor transport technique, although the charge sometimes consisted of separate ampoules of metal and phosphorus heated to nearly ~he same temperatures. I~ow-ever, the flow of vapor species to the deposition zones was eiEectively the same as when the metal and the phosphorus are first mixed together More speeifically, in single source vapor transport the vapor species are first bxought together at a high temperature and then are deposited at a lower temperature.
The following indicates our development of the techni-que as it has been applied to the preparation of alkali metal polyphosphides and the departure from von Schnering's method, which results in improved, more selective prepara-tion of: crystalline metal pol~phosphides of the type KP15; low alkali-metal content polyphosphides, polycrystalline material, of the type ~Px, where x is much greater than 15; and a new form of amorphous phosphorus~ in which the ~Xalinetal content can be less than 50 ppm ~parts per million).
~ he studies we have made fall i~to severa~
categories: type of charge, char~e ratio, tube length and geometry, and temperature gradient profileO The following examples illustrate the temperature dependent product deposition relationships we have discovered and our improve~
temperature controlling methods that result in the selective preparation oiE desired produ~ts.

~40- 110-001 General Methods:
__ __.

An alkali metal ~nd red phosphorus are.sealed ir, quartz tubes, at reduced pressureC (about 10 4 Torr) r Ato~ :;
ratios of the two elements range from P/M = 5/1 to 30/1, .1 with lS to 1 as the most common charge. The el~ments are generally ball-milled together, prior ~o loading in the quartz tubes~ The millings are carried out with stainless steel balls and mills and last for at least 40 hours. The ~, mills are usually heated to 100C for the duration of the ~.
milling, to assist in the dispersion of the metal in the red phosphorus powder. --The milling achieves an intimate contact of the two elements in as homogeneous a manner as possible~ The prod~
ucts of the milling are generally fine powders which are easily manipulated in a dry box and may be stored with little noticeable deterioration, The powders show remark- .
able stability when exposed to air and moisture, compared to the stability of their constituents, especially the alkali metals. For instance, direct addition of water to the pow- .~a ders only results in combustion of material~ in random ca~es :-~
and on a small scale. ~ .
.
Preparation of MP~ single crystals, polycrystalline ~.:
and amorphous mat ial~ u .
: ,....
A mixture of the elements (alkali metal and red phos-phorus~ i5 sealed at reduced pressure lless than 10 4 ~- -Torr) in a quartz tube 58 (Figure 3), about 50 cm long by .
2.3 cm in diameter. Tube 58 is supported inside the heating -~
chamber of a hindberg Model 24357 3-zone furnace in one of two ways. One method employs a second quartz tube 60 as a : :~
support piece, which is, in turn, held in ~he chamber, away from the heating elemaAts, by asbestos blocks 68 and 70 such that the coupled tube~ rest at an incline, insuring the reactants rem,ain in the hottest zone~ The other method .~:
~Figure 14) :is to use supports built o woven tapè137,139 ~:
wrapp~d a~out the reaction tube'in an expanding spiral., an i~ch wide, a~ld filling the circular cross section of the *Trademark ~12~ Z~

-41- llO-OOlB

heating chamber. This woven tape may be made of a variety of materials: Asbestos, Fiberfrax (from Carborundum Company), or woven-glass. The :Latter is preferred primarily on safety and performance criteria~ The implications of usin~ the two different methods are described below.
The reactants are driven to products by applying energy to the system via the resistan~ce elements of the furnace.
If a sufficie~tly high temperature is applied to the re- :
actants t while other portions of the tube are held at appro- ;
priate lowar temperatures, products will deposit, or con-dense, from vapor species. The temperature differential which drives this so-called "vapor transport" synthesis, is -~
achieved in a 3-zone ~urnace by selecting different setpoint temperatures for- th~ individually controlled heating `~ -elements. ~-~
~ T~OD 1. See Fi~ure 3. The 50 cm tube~ containing the reactants,is held by the second quartz tube in the 61 cm long heating chamber. Application of a thermal gradi~nt b~
manipulation of the 3 set-points results in a generally linearly-falling gradient. That is, the slope of the grad- ~ ~
ient, ~ T/d, wher~ ~ is temperature and d is distance along - -.3 the chamber, is approximately constant between the centers ~;~
of the two outside heating elements. This linear gradient, applied over the long dimensions of the tube, functions to cleanly separate the variety of product materials formed in the reaction. The products occur in a characteristic -~-~
pattern of decreasi~g temperature of deposition: dark purple to black polycrystalline filmst a ring ~f massed crystallites; "single" crystals or whiskers; red films of small-grain, polyory~talline morphology; and~ at coldest temperatures, dark grey amorphous material.
A serie of experiments have shown that the amorphous material will not form in these sealed tubes if th~ coldest temperature is greater than about 375C. Similarlyp the occurrence of the red polycrystalline material could ~e greatly reduced by keeping the lowest temperatures at or above 450nc.
.
* Trademark We have also found that polycrystalline MP15 will not form in single source apparatus~ The polycrystalline and amorphous materials forme~ are all high ~ materials where x is much greater than 15.
METHOD 2. The woven tape holders serve no~ only to ori~nt the reaction tube but also as efective barriers to heat-transfer between the three heating zones. These barriers give rise to steeper drops between the zones, but a flatter gradient within the center zone. The result is a step-like temperature profile, which can be manipulated to selectively produce products by providing appropriate ranges of deposition temperatures.

~ Determ;nation of Product Deposition Temperatures .

In von Schnering's announcement of the preparation of single crystals (whiskers~ of KP15, he described the preparation Erom the elements as entailing the heating of the elements--potassium and red phosphorus--in a "temperature gradient" of "600/200C", in a 20 cm or so quartz tube. He further states the crystals form at "300 ~o 320C"~ The furnaces used were apparently single element furnaces in which the gradient arises via heat loss from one end of th~ tubes sticking out of the furnace.
In the first improvement on this procedure, a three-zone furnace as shown in Figure 3, with i~dependently controlled heating elements~ and a 61 cm long heating cham-ber ~Lindberg ~odel 54357 3-zone furnace) was.employed, to achieve and control the applied gradient. By supporting the reaction tube, which was now extended to approximately 52 cm long, in a second t open ~uartz tube, which was, in turn, supported by asbestos blocks, a generally linear temperature yradient, ~ T/d, was approximately constant ~etween the centers of the two outside heating elements~ The power to these elements were controlled by a Lindberg Model 59744-A
Control Console, which uses three, independent SCR-propor-tional band controllers to maintain the temperatures seleeted on manually set thumb-wheels~ .
.

*Trademark.

~1 . j 12~
~43~ llO-OOlB

The linearly-falling gradient, applied over the long dimensions of ~he reaction tube, served to cleanly separate the variety of product materia:Ls formed in the reaction.
The products occur in a characteristic pattern o~ decreasiny temperature of deposition: dark purple to black polycrystal-line films; a ring of massed crystallites; single cry~tals or "whiskers"; red films of small-grain, polycrystalline morphology, and, at the coldest temperatures, dark grey, amorphou~ materials.

Example I
A Lindberg Model 54357 3-zone furnaee as shown in Figure 3 comprising heating elements embedded in a refra~-tory material in separate cylindrical sections of 15.3 cm, 30.5 cm and 15.3 cm lengths, for a total heating chamber ~ -length of 61 cm,was used for this example. The diameter of the chamber is 8 cm. Controlling thermocouple ~not shown) ~
are located at about 7.0, 30.5, and 53.5 ~m along the 61 cm ~:.
length.
The ends of the heating chamber were plugged with glass ; :.
wool to mi~imize heat loss from the furnace. A 60 cm lo~g by ~:
4.5 cm diameter quartz tube was held at a slight angle, by ^-asbestos blocks, in the heating ch~mberO
The quartz reaction tube was round bottomed~ 49 cm long s by 2.5 cm in diameter t and reduced to a narrow addition tube 10 ~m long by 1.0 cm wide, Under a dry nitrogen atmosphere, ::

6.51 g o~ red phosphoxus and 0.62 g o~ potas~ium were trans- -ferred into the tube~ The atom to atom ratio of phosphorus to metal was 13.3 to lo The phosphorus was reagent grad~ ~ :
tJ.T. BaX2r). The tube was evacuated to 10 4 Toxr and sealed by fusing the ~ddition tube several cm'~ from the .:
wider part of 'the tube such that the total length was Sl.5 cm. The seale!d tube wa~ placed in the 3-zone fuxnace as j.
described above and the set point temperatures of the three zones brought to 650C~ ~50C, and 300C over a period of 5 .:.
hours, and heldl there for.another 164 hours. The power was _44_ 12~S2~ llo OOlB

shut off and the oven allowed to cool to ambient tempera-tures at tlle inherent cooling rate of the furnaceD The tube was cut open under a dry nitrogen atmosphere in a glove bag The products consisted of crystalline, polycrystalline, and amorphous ~orms.
In Ta~le III, the different processing parameters used for several other runs are listed, along with the types of products observed in each run. Prior to being cut open, the tubes from the first three runs were inspected as to the po-sitions along the tubes of several products: the dark ring of massed crystallites, and the start of red polycrystalline films. The whiskers were always observed between these two points. These positions were later correlated to the tem-peratures along the gradients created by the noted set points. These data are recorded in Table IV.

- 4 5 - 11 0-0 0 lB
2~
~n X I H ra s-~
d ~ C~ Q r E~ X r-l ~ .rl Q~ ~ ~ E~ fd ~; ~H

~O
U
rO P.
~o--~ ~ o a S-l V ~.) ~ VC,) ~ V
u~
,: O O
X ~ ~ L:
U ) O C~ ~ ~r l 1~ 0 u~
,~ h U~ at t~l 1-- ul ~5~ a) Q) Il) ~) C) ~ ~ ~ ~ ~ ~o ~ s~
~1 0 E~-- ~ c~ o ~ o ~ tJ~ O
S~ ~ O--u) O o o u~ oa) h ~ r~ u ~1 ~,) O U-) O 1~Ll') ~ 5 J ~r l ,~
o ~ ~ O H h ~1 ~
1~ n5 a) X O
o u In O Oa) o Q~
H N ~ 11~ r l H E- D ~ ~r ~ ~ ~~ ~ ~ ~ 1-1 E3 o o o o o ~ ~ ~ ~
~ r--l ~,) 1~ 0U') O O t~ r~ ~
m E~ D ~ Ul O r~l f~ .r~
' , I I I I I ~ O rd In O O O O O ~ O
~ O -- X ~ X X X U~ ,!c r--P E-~ U) r~ ~ I~ r~ r~l ~r ~o O ~ ~ 3 a) E~ r~ D O~ ~r~
p. ~ ~ o ço o~ ~ ~ ~I h h ~ U~ ~ -r l ~ O ~ ~r l ,~

U~ r~

u) K K K ~ 0 R 4i -- ~ -- z z ~; o h ~
I ~ ~ ~ r-~ .~ h ~ u~ ~:
O O O D ~U Q) ~I) >'~ O h o o o u~
r-J r~
~rl ~ ~ 10 ~1 0 0 . Ia a) O ~rl O
~) ~1 Ir) u~l o O ~1 0 J~
.C P; r~¦ r-1 ri r~
.) . U~
~ a~ h r a) o .,~
Z r--l N ~ q~ lrt ld ,S~ O

- 4 6- llO-OOlB

L~
C~ O E~ o o o o o o Lr O~ .
'.c a~ o cr ~ ~ E~ E~ E 6 ~ O
o ~n o u~
o a~ ~ ~ . . . . ~ .
W ~ ~ ~ 7 ~ O ' ~ ~ o ~Z :' C~ C~ U t~ o o ~ - ' .,1 ~ o o o o o ~ ~, .
6 ~n o o u~ o a~
~r er , U~
~ u c~
~ o o u~ In o Ul 1~
Q~ ,1 '~ . . . . O ,3 ~4 O 1'1 ~57 ~r N C) rl al nl E~
o ~ u ~ ~a ~ a) -~
.1~ QJ o o o o o ~:4 h U ::
a~ ~ ~ O
v) ~ t-- ~ o_~ ~ t.q a.~ ~ o E~ ~ ~ u~ o ~ O O
H ,~ +I
O ~ O -, :, ~::1 ~ O ~: O If ~ U) O U~ _I O ' ~, q ~ P: ': ''-':
~: a~ O
:~ s~
h ~1 ~ ~ i u ~d 0 ~ ~, .. ...
~ O u7 d) o o o o o t~7 O U~ O O O
- r~ ~ el' ~ ~ ~A ~r/ ~ ~rl ~ :
o ~
OU~
x ~. - ~ ~ 3 0 .~ 3 ~
:~ O o o o O U~
o~ o ~ ~ ;-t: ~1 ~ ~3 ~ Q) d h q~ --~1 ,1 O ~ Id o ~ ~ ~. 3 ~
O P. u~
P- a~ d O
r ~ ~ rl : ;
O O
Q~ O , Sl S~

-47- 121~ 5~1 llO-OOlB

The information from these two tables was used to es-tablish the relationship of temp~rature and product-type.
The single crystals of KPl~ appear to form over a temperature range of about 40 ~ 10C, the center of which varies from run to run, but wh:ioh lies around 465-475C.
Similarly, the onset of deposition of red, polycrystalline materials appears to be about 450 + 10C. Finally,amorphous material deposited even when the lowest temperature was around 350C. When this was raised to 400C, no amorphous materia~ was observed. (Although the run in which this temperature was used eventuall~ ended in a failure of the reaction tube, before products could actually be harvested, this temperature-product relationship for khe ~norphous -;
material was confirmed in later runs using more advanced techniques). Assuming a mid-range value~ an upper limit fox deposition of amorphous material was taken as about 375C. -The pressures in the heated tubes were not measured~

B. Temperature Gradients Which Favor Growth of Single Crystals ~Whiskers?
, ?
Using the knowledge of the deposition temperature- ~, product morphology relationships of Tables III and IV, im- -pro~ements in the synthetic technique were sought which ~ ;
would allow greater selectivity of product type. Methods were sought for manipulating the temperature profiles in the - ~
fur~aees which would result in larger areas of the tube sur - J
face being within the appropriate temperature xanges for -gi~en products. Several available materials with low t~er-mal conductivities, and in easily manipulatable forms were checked for u~e as barriers to heat transfer in the fur-naces. Woven apes of asbestos proved a suitable product for both supporting the reac~ion tubes and creating complex gradients, consisting of areas of fairly flat, or isother- ~ ~-mal, temperatures, separated by areas ~cross the barriers~
of steep drops or gradients. These so-called ~step-like profiles were applied in all the subsequent examples wher~
specific product:s were being sought in maximum yields, . . , "' -48~ 2i l10-OOlB

Another improv~ment which helped gain more reproducible temperature profiles from run to run was to use a more solid, cer~mic type of material to fill the gaps in the heating chamber walls In early runs~ these were plugged with glass wool, which helped stem loss of heat, but not very efficiently. The large cylindrical gaps are p~sent in the chamber walls because the furnaces are actually designed to hold a process tube along it:s length, for flow-thxough applications, rather than for enclosed systems, as are ~eing run in these methodologies~
The following examples wer~ all aimed at trying to pro-mote growth of single crystals, both larger in size, and in greater yields, both as a percentage of product forms, and in absolute yields. These results were indeed achieved.

Example II
A Lindberg Model 54357 3-zone furnace identical in de-sign and siæe as that of Example I was also used in this example. The elements ~ere likewise driven by the same man-ually set model 59744-A Control Console. The ends of the heating chamber were plugged with a heat resistant ceramic-like material, to minimize heat loss from the furnace. The reaction tube was supported in the heating chamber by two rings of woven tape of asbestos. One of these was located between 16-19 cm and the other between 42 and 45 cm along the chamber. This put both rings completely inside the center heating section, just beside the junctions of the center elements and those of the two outer sections. The rings were constructed such that the tube was held at a slight angle. The ring served to insulate the heating zones from each other by acting as barriers to heat transfer.
The quartz reaction tube (Figure 3) was round bottomed, 48 cm long by 205 cm in diameter, and reduced tv a narrow addition tube 162, 10 cm long by 1.0 cm wide. Under a dry nitrogen atmosplere, 5.47 g of red phosphorus and 0.50 g of potassium were t:ransferred into the tube. The atom to atom ratio of phosphorus to metal was 15~1. The phosphorus was 99.9999% pure. The potassium was 99.95~ pure, The tube was ~49~ 1~1~2~ llo OOlB

evacuated to 10 4 Torr and sealed by fusing the addition tube several cm's from the wider part of the tube such that the total length W25 52 cm. The sealed tube was placed in the 3-zone furnace as described above and the set point temperaturcs o~ the three zones brought to 600C, 475C and 450C over a period of 4 hours, and held there for another 76 hours. The power was shut off to all three zones at once and the oven allowed to cool to ambient temperatures at the inherent cooling rate of the furnace. The tube was cut open under a dry nitrogen atmosphere in a glove bag. The products consisted of crystalline and polycrystalline forms.
Table V lists the processing parameters for a number of other such runs (data for the above example are from the run with reference number 10~.

Z~L l l O - O O lB

Cl~ o S
~ ~ ~ o oo o ~ ~
R ~ O O ~ O O
~ a) e o R R t`J ~`1 ~ ~ o ,q Q

~D~D er O ~ o o CO ~ ~ o~
~ ~ ~ ~ ~ ~r ~~ o~r 'rI E-~ 'O 'r ~ D ~~ N t`~ Q
OOOOOOOOUlO
E~ o ~ e~ ~ ~ el~ ~ el~ ~ t~l ~
u~ In O O U~ O O

o o o 9 o oo o c:~ o ~ o o o o o oo U~ o o E~
d~ U~
a~ O
. o o o o o oo o o o cn ,~:

h o o x x x x x xx x ~e x ~ ~1 pl ~
ul R Q ~P
O ~ ~ ~ D O ~
Id P~ co o o o~ ~o ~ I~ ~1 00 o~ 1:
a~
U~
`5 ._ _ ~ K
u~ Q .a --`~
c~ O O u) r- OD O ~ ~I ~H
~ ~ ~ ~ o C~ O _I ~1 0 0 0 0 0 0 0 ~ I
.~

hrl ,1 ~ co ~ ~ ) O ~
h O
O ~
~ ~ 0 O O ~ ~
5~; ~,1 h~,l 13.C R-~
al o ~r~

-51- llO-OOlB

All of thcse runs rcsulted in crystalline and polycrystalline ~orms. The yields of the single crystals were always greater than in Example I. The polycrystalline materials were always in the form of films deposited in the colder ends of the tubes and were usually limited to the last 10 or so cm of the tu~e, though there was usually some overlap with the single cryst:als. Single crystals from these ru~s were characterized by X-ray powder diffraction patterns as having the same structure as KP15 as determined from XRD data. Wet chemical analysis of the crystals were difficult to o~tain with great accuracy~ i~
part because of their stability, which required extreme conditions for digesting the materials for analysis. (See the tables VIII through XI on analytical data below) The polycrystalline films were also characterized by X-ray powder diffraction methods and wet methods. The ~ilms showed varying degrees of crystallinity, and the patterns were similar in several aspects to that of KP15, but yet were disti~ctly different in others. Furthermore, the wet analysis, coupled with flame emission spectroscopy consis-ten~ly showed the alkali ~etal content to be in the part per million range (i,e. less than 1000 ppm and often les~ than 500 ppm3, and with P~K ratios ranging from about 200 to 1 to about 5000 to 1.
, ~. ,, C. Thermal Gradien~s Which Favor Growth of Polycrystalline and AmorphDus Materials ~ ollowing the successful improvements in production of single crystal materials, a similar series of experim2~ts was carried out to manipulate the 3-zone furnace and asbes-tos rings to find tha stepped thermal gradients appropriate to selectiv~ly produce the polycrystalline and amorphous materials observed in earlier runs.
These eaxlier runs suggested the temperatures necessary for obtaining the desired products. What remained to be shown was how to optimize these products. Table VI shows t~e type of prQ~iles used and the products observed, ..

~ '~

-52~ L~2i llO-OOlB

~_ ~ Q~ ~ Q) ,a ~o ~
O O ~1 0 R J ~) ~1 O
X CJ' x x ~ x x ~ .
~4 ~ X X X X X X

o U~ ~ o o o ) ~ O
l ~/ o R O ~r 1` r--U~ ~D ~D 1`-~ ~ ~ a~
E~ ~ ~
~1 0 tD N N N C~ o N

/~ ~ o In In ~'V tt~ 11~ U'~ N N
E~ o r~ ~ ~ ~ N ~
H ~ 0 in 0 1~1 0 :- t~V ~D N 1` u~) N
E '~ ~ ~ r~l ~ t~
~3 O O O C:~ ~ O
~ {) O O O O O O
E-l E- O
er ~ L~

C~ O O ~ O O
tq r-l ~ r~l _I r-l r-l h U~
0 S~_ X X ~ X X X
O~S
~-~
~q 5~ Q~
E~ ~~ ~ o o L~ U E
flS ~ r o o~
o ~ Ul I ~I r-l Q a~

Ul I ~ X ~1 ~ ~ rl r l O t'~ 1 ~1 0 X 4~
U) ~ ~U) ~ O E~ ~ O ~ ~r~ rcJ
~1 ~ U7 rt 0 ~ 0 IU ~ Ql O r^l r-l~I r~ C ~ -r~
h r~l Ql 0 ~ r~ i~l Ql t~ O Ll~
rl r-~ ~1 ~ Lt') u~ _I O rl ~3 .SJ ~ ~ ~ Ul O
dP~ ~ ~ K p~ h a~ O ~
' ~ ) U ~ rS
U~ ~rl r I O
O . O
~ ~ , Q~ . ~ ~ Ct~

The first run, which is the subject of the Example III, just duplicated the temperatures of the ranges used in Example I, the linear falling gradients now changed to a stepped gradient. Not surprisingly, all product types were found, with some variation in quantity, compared to those of section A. When the coldest temperature was raised to 400°C, as in the second run of Table VI, no amorphous material was found, as anticipated. With the 425°C center-section temperature, however, nearly two-thirds of the tube's interior was covered with polycrystalline films, and only a small number of whiskers were found, meaning the films could be produced almost exclusively.
In the third and fourth runs, though, where the coldest temperatures were held at 350°C (cold enough for amorphous material to be formed in the first run), and the center zone temperatures were lowered to 375 and 350°C, the amorphous materials were no formed in large amounts at all. Instead, large amounts of both single crystals and polycrystalline material were found over a fairly short space of the tube, and at best, only think films of amorphous materials may have formed in the rest of the tubes. The same phenomenon was observed in the next two runs as well, although there were definitely think amorphous films in one run. Apparently most vapor species are condensed out in the polycrystalline and single crystal forms and no significant vapor travels to the region which is cold enough to form amorphous forms.
Example III
A Lindberg Model 54357 3-zone furnace identical in design and size as that of Example I, was also used in this example. The elements were likewise driven by the same manually set Lindberg Model 59744-A Control Console. The ends of the heating chamber were plugged with heat resistant material to minimize heat loss from the furnace. The reaction tube was supported by two rings of woven asbestos tape. One of the rings was located between 16-19 cm and the other between 42 and 45 cm along the chamber. This puts ;52~
-54- llO-OOlB

both rings completely inside the ce~ter heating zone, just beside the junctions of the ce~ter elements with those of the two outer sections. The rings were constructed ~uch that the tube wa~ held at arl angle. The rin~s also served to insulate the heating zones from each other, by acting as barrisrs to heat ~ransfer.
- The quartæ reaction tube was round bottomed, 48 cm lon~
by 2.5 cm in diameter, and reducecl to a narrow addition tube 10 cm long by 1.0 cm wide. Urder a dry nitrogen atmosphere, 5.93 g of red phosphoru3 and O.50 g of pota99ium wexe transferred into the tu~e. The atom ratio of phosphorus to metal was 15. The phosphorus was 99.9999~ pure. ~he potassium was 99.95~ pure. The tube was evacuated to 3 x 10 4 Torr and sealed by fusing the addition tube several ~m's from the wider part of the tube such that the total length was 51 cm. The sealed tube was placed in the 3-zone furnace a~ described a~ove. The temperature gradient was driven to 600C, 465C and 350C over a period of hours a~d held there for 7~ hours. Th~ power to the elements was then shut off simultanaously and the furnace allowed to cool to ambient temperature~ at the inherent cooling rate of the urnace. The tube wa~ cu~ open under a dry nitrogen atmos~
phere in a glove bag. The products consisted of single crystal~, poly~rystalline films,and amorphous m~terial.
.... ~''J
D. Production o Cylindrical Boules o~
Amor~ou~ Polyphosphides ,~ ~
~t wa~ e~ident fron~ the experime~ts describe~ in ~ec- - ¦
~ion C that to obtain large amounts of amorphous materlal improvements ne~ded to be made in the processes already being u~ed. It wa~ recognized that in order to gQt bu~k ~onms of th~ material, as opposed to thin films, the con~
ditions appropriate for ~rowth had to be conf~ned to a smaller ~pace th,an previously allowed. This tran~latea ~nto - ~ ~
allowi~g onl~ th~ extreme end of the tube to be at ox below . ~ i 375-C or so. ThLs was ~ccomplishable in principle by use of ' I

'Z ~

-55- ~ % ~ ~ Zl llO-OOlB

the thermal barriers. However, it was also recog~zed that if the conditions for formation of other materials, i.e.
single crystalline MP15 ox polycrystalli~e MPX (x is much greater than 15), were also ,availahle over a large area of the tubc, these materials would act as "traps" for vapor spe~ies. It was therefore, also necessary to discourage the fonmation of the other materials, This was accomplished by raising the center zone temperatures to levels which would be too high for formation of polycrystalline or single crystals. The only area the~ where these materials were favored were through the area of the thermal barrier, where rapid temperature drops occurred~
A3 shown by the following example, and other experi~
ments summarized in Table VII below, further impr~vements in the procedure were worked out. The first was the use of ~oneywell Model DCP7000 Digital Control Programmers to drive the heating elements. This allowed the pre-programMing of the temperature changes sueh that reFroducible ~reatments could be made from run to run. Both controlled heat-ups and cool-downs could be accomplished, eliminating tube failures, and productio~ of white phosphorus, The latter often occur-red when tubes were cosled rapidly and phosphorus vapor con~
densed a~ P4. This was often the reason materials appeare~ reactive. This reactivity could ofte~ be removed by soaking the materials in solvents which would dissolve away the white phosphorus. The ~econd improvem~nt was thQ
routine of applying an "inverted gradient" of 300-490-500C
across the tube ~rom th~ meta~/phosphorus source to the depo3ition zones before vapor transport, which cleared the depo~ition zones of materials, which might affect nucleation processes~
By ar, t~he mo5t important improvement, however~ was redesigning the ~eometr~ of the. tube. Instead of a lon~
tube of nearly uniform 2.5 cm diametex, the body of the tube was shortened to about 30-32 cm and the 10 mm diameter addition tube 160 ~Figure 2~ lengthened and sealed such that about S-7 cm of this tube remained as available space in the interior of the~ tube- Wh2n t~is latte~ section was pl~ced -56- llO-OOlB

in zone 3, and the vapor ltransport gradient applied, this section bPcame filled with solid, bul}cy cylinders of in-crea~sing len~th, as the conditions 'or ~rowth were improved.

Exan~le I~l P~ Lindberg Model 543S7 3-zone furnace, i~er~tical in design and size as that of Example I wa~ also us~d in thi~3 exampl~. The elements, however, were driven by a Honeywell Model DCP-7700 Digital Control Programmer which enabled processing to be pre-programmed and carried out in a repro-ducible fa~hion.
The ends of the heating chamber were plugged with heat resistant material to minimize heat loss from the furnace.
The reactio~ tube wa~ supported by two rings of asbe~tos tape. The rings were constructed such that the tube wa~ .
held at a slight ang~e. T~e rin~s also served to insulate the heating zone~ from each o~her.
The quart2 r~action tube was round bottomed, 33 cm long by 2.5 ~m ~n diameter, and reduced to a narrow addition tube 162, 20 cm long by 1.0 cm wide. ~nder a dry nitrogen a~mo~-phere, 7.92 g of a ball mill~d charge of atom to atom ratio of 15 to 1 was ~oaded into the tube which was evacuated to 1 x 10 4 Torr and sealed by fusing the addition tube 10 ~m from ~he wider part ~uch that the total length was 43 cm lon~7. The seale~ tube was placed in the 3-zone furllace using th~ woven barriers described above.
With the tube between 6 and 49 cm, one thermal barrier a~ 16-19 cm and th~ other at about 38-40 cm, the ~oneywell P~ogrammer was used to apply an "inv~rted ~ra~ient" of 300, 490t 500C fox 10 hoursO After the furnace cooled at th~
inherent ra~e o f the furnace, the tube was moved to li~
~etween 12 and 55 cm. The thermal barriPrs were also r~--arranged to lie at 18.5-21.0 cm and 44.5-47 cm. The pro-grammer then drove the gradient to 600o~.485oo 300C or 64 hours. The programmer then took the tube through a con-~rc>lled c:ool-down ~3e~uence to a 180t 190, ~OOVC ~radien~

*Trademark ,~ !
~ .

~57~ llO-OOlB

which was held for 4 hours. The furnace was then allowed to cool to am~ient temperatures at the inherent cooling rate of the furnace~
The tube was cut open under a dry nitrogen atmosphere and 4.13 grams of a 2-3 cm long solid homogeneous amorphous boule recovered from the addition tube 162 ~Figure 3).
The results of several other runs are shown in Table VII.

"-.

~'`'':, "~

- I

-58~ llO-OOlB
5;~3L

~a u~ a~
o .,~
e e au e a~
~J ~ O
O ~ ~ ~ ~ O ~ ~ dP ~ d e o ~ ~ o o . o~
S

a~ ~
Q ~ ~ o o o o o o o o 1~ In ~ U . . . . . o E~ ~ i` a~ o~ o r- ~ ~_i ~7 In C~ OD 0 ~ ~ r~
~1 0 ~ ~ ~ ~ o o o o C~ o U7 o o o o o o o ~ o ~ ~) ~ o o o o o o o o o E~ o ~ ~ ~ ~r~

H o O u~ lL~ O u~
t )~ OD CC\ t~ ~ O CC~
E~ o ~r ~P ~ ~r ~ ~ eP~r u~
~3 E~ o o o o o oc~ o o o -~ C.)O O O O O O O O O O
o U~

O O O O O O O O O O
10 Sl &) ~ X X X X X X X X X X
h O

Lt'~ ~11'1 ~ N ~ 1 o r~ ~ o ~ ~~a o ~
. . . . ~ . . . . ., n R~
m r~ m m ~ ~ m ~ ~ ~ ~ ~ ~ l-- ~ ~ _I ~
i~ ~ X ~ ~ X ~
Z .
41~
a~ --' ~ ~7 ~r ul ~ 1` C~ C

-59- llO-OOlB

U~
o .c Q) P~ ~ dP A ~P
o ~ t~
O dP' dP ~P . . . . .
~D ~1 ~ ~ ~ ~ r~ u~
~C
Q) ~
o ~ o ~ U V~ o o ~ . .
U) ~ U~ U~
r~ er ~ ~ ~ r U~
~3 ~ ~ ~ ~r ~ er ~r ~
.~ o o o o o o o o o ~1 ~ ~ ~ ~ ~ ~ ~

D O O C:> O O O O O
:Z t~ U O O O C:> O O O O
H E o ~ ,r~ ~ t'~ ~ t'~ 1~) r'l U

O O O O O O O O
Ht~l U O O o o O O O O
HE-l o u)u~
W
C~ O ~ C~ O O O O
~ ~1 ~,) O O O O O o o o E~ E-~ o~ ~ ~~D ~D ~ tD ~D
*
OO O OC:~ O O O . t) U~ ~ ~ 0 CJ ~ XX X X X X X ~ S
O d U~
~oo ~ oc~ co~n~ ~ ~ 4 ~0 r-- ~ ~In _I ~ ~ ~` o o ~ 41 .. . . ~ . . . ~ O O
' ~ O id ~: æ :: o w h ~:~ ~P. mP~ m . m c~ ~ U u U~ u 1~> s .s::
s ~~ P~ ~,a ~ ~ ~ Q) Q~ --I a ~:~C K ~ ~ X æ u ~ o~
:~
~ ~
O
Z ~ 0 0 ~
,_ _ a~ o,~~ ~ ~ . u~ ~ 1-- ,. p:, :~ `
c~ ~ '~ ~ ~ ~ ~ ~ ~ ,~ v m ~
, -60- 121~5~1 llO-OOlB

The results showed the yields of material to be fairly independent of the char~e type - i.e r ball milled, or ~he pre-reacted condensed phase product5~ However, there was a distinct dependency of yield on the P/~5 ratio~ The greater the relative amount of metal in ~he charge, the lower the yield of material. As the amorphous material is essentially phosphorus, this reflects a lower vapor pressure of phos-phorus over a metal-phosphorus charge the greater the metal content, hence, a slower rate of growth for identical thermal conditions.
Table VIII conta~ns some analytical results on amor-phous boules prepared. It shows potassium content, as de-termined by wet methods. It also shows trace constituents shown to be present by Flame Emission Spectroscopy~

-61~ 5Z~ llO-OOlB

U) a .~ ~
~: ~ o o W ~D
U~ ~
'~ ~ s o Ut W ~ U~ o _I o U~ o ~ o C~
~ '~--I ~D U~ ~1 ~ ~r r~
~r~ Q ~ Q I I I I o Q~ ~ ~ O
o .. ..
:~ U I 1~ z u~
at S~ o h h `lJ
u~ ~ a) ~ o o qJ ~ ~ ~ ~
O ~ ~ .. ~q Q~ ~ ,X
U o o o ,~
O C O ,C,~: O N C I ra H ~ Q) ~ O ~J O U ~ 1;) ~ ~1 H~1; ~ ~d ~ ~ I t~
H ~ ~ E~ I I OI ~ ~) ~ O 5~ 0 X ~W ~ ~ ~ ~ ~ ~ O
rc~ ~ r r~
O ~1 ~ ~
~: O ~ ~1 E4~C u:l 1¢ ~ 1~ ~ ~C ~ K Q) r-l ~ Id :S l~J U~
.
u~ :~ Q. ~`1 Ul .4 ~
O ~ r~ In o o OD
C~ ~1 ~ co ca O
~) 0~
E~ ~1 u~

.;:~ ~ ~ ~ ~ ~ O P~ O
K ~ K ~ K S~ .l O U~
a tn lo ~ ,4 -,~
, -~1 a~ ~1 Z
m~
P; ~J

--62- ~2~L~5~ llO~OOlB

Tables IX, X and XI are of analytical data obtained b~
wet methods on product from vapGr trans~ort synthesis.
The P/M ratios in the tables are atom ratios unless otherwise noted.

TABLE IX
SINGLE CRYSTALS [~HISKERS) FROM VAPOR TRANSPORT
Ref. No. _Charge P/M Total*
38 K/P15 lg.l 94.5 6 K/P15 19~1 98.8 K/P15 19.1 99,4 11 K~P3~ 16~4 96.1 17 X/P5 11.3 g7.7 * Analytical mass balance ~M + %P detected .

-63~ llO-OOlB

TABLE ~
_ . . .
AMORPHOUS MATERIALS FROM VAPOR ~RANSP~RT
Ref~ No. Char~ P/M Total*
3g K~P15 2500 W 100.3 16 ~/P15 ~'750 W 99.7 21 K/P15 2300 W g2.8 22 X/P15 ~.~200 W ~7.0 ~5 K/P7 3500 W 97.9 26 K/P15 6200 W 97.8 23 R/P15 greater th~n 98.2 ~500 W
24 K/P15 7000 W 93.3 24 K/P15 25000 W 99.5 27 K/P5 greater than 99.7 84~00 E
28 K/P15 7800 W 98.2 29 K/P15 25000 ~ 94,~

~) Wet analysis E) Flame emission spectroscopy * Analytical mass balance %M ~ %P detected -64 llO-OOlB
~, ' .
1 ~ O ~ ~ G O O

~.1 10 ~'1 O O ~ r_ C~ ~ ~ N
>~
~ J O
_I P. ~ O ~ 1: 0 0 1~ _1 ~.1 ,C U
O O O~ .. 1 r_ o ~ ~ V
- h .n o~ a~ ~ _I w c ~; ~ r~ r o . Ia ~ o o ~ o o o o O _I * 'I s: N ~ ~ ~_V o ul~ o Cl~ o o o 1~4_1 lh .~~ N ~ N
E~ ~. " ~ ~ oo~
D. O ~ tn 0~o~ O~r' O
X ~ _l I ~ ~ o ~ o U~
C ~: ~ P.
~ ~ (1~ ,~
o 2 ~ ~ o ~ o E ~ Il. r o ,~ ", O

:~ O U~ D C
0 ~ C:- O 1~ I`~1` ~ O
e~ N C'~ > ~ E
8: ' r~ ~r 0, ~ t~ a~ V
o ~ ~ U~ o o~
~ u~ IJ
0 ~ m N u~ U~ o ul o~n ul . U~ ~ u~) ~
D~=~yYYYY :c:C Ic X
O ' O~ ~O O 1~ ~D ~ OO~ O ~ N u~ o 1~ Z r~ N ~ ~N~ ~`11 ~ N ~

-65- llO-OOlB

,Y. C~ o ~Po ~ ~ ~ V
, ~ ' ' ~ u 1 v ~ r~~ rs 0 0 ~: ~ D'Z ~ Z 'O Z

G~ OG~ O d~
o ~ ~ o e~ o o c a o O a~
E~ o~
U~ ~ O
::~ ~ ~ o ~ ~ o~
O_~G O h ~ O ~ 1: 0 ~ .C O ' ~ O
t~.~ _~ tPV r~ t~ ~ E
Z OO O ~ N ,~ ~" ,q O
O ~Sl~ ' ~ ~ W ~ ~ h ~ O1~~ ~ ) 1 o O C ~ O ~ ~: a t O O ~ O O ol C~
~A ~ ~ ~ ~ h ~ N U7 ~ O ~.1 .C h ' O O ~ ~ Ul O 1.1 W ~t:~ .-1 ~ V 0~ ~D V ~ ~4 D' V ~ ~ ~ ~ ~J r~ I h :~~1:~:~ 0 N U~
~: ~4 O o ~ w j~
~ ~ E-~ o~ 17 0 t~ ~ C~
1.0 '~

~ ~ ' ab !~ ~ e',aJ
a~ 0 ~ E
:- ~ ~ . ~ ~., ~ b~

o o ~ ~ o ~
~ o~ ~ ~ D Ul ~

0 ~ ~ X ~C D z Z 7 ~ i D--N t~ ~ ~ ~, ~1 ~ ~ U~ 3 _ IL~2~
-66- llO-OOlB

~rcpaxat~on of Metal Polyphosphides by Two Source Techniques P~lyphosphides have beer, prepared in two fundamental].y .
different types of equipment which are both identified herein as Two Source or separated source techniques because in both types of equipment, the metal and phosphoru.s are separated a~d heated independently on either side of a deposition ~one. All examples hiave been carried out on the K-P system. :~
In the first method, as shown in Figure 11, the phos-phorus and potassium charges are held at opposite ends o~ a sealed quartz tube 100. The tube is subjected to a temper- :.
ature profile as shown in Figure 12, achieved by use of a three zone furnace. The profile take5 the indepe"dent ~:
charge5 to elevated temperatures, relative to the center zone between th~ two constituents. In this zone, the vapor- ~
ized constituents combine to form the deposited product of ~.-KP15, in the form of films o~ the reactor walls. (More com- .
plete details appear in Example V below) In the secon~ apparatus, a~ illustrated in Figure 14, a su;:s~tantial section gener~lly indicated at 102 i~ at a~ierlt :
temperature held outside the three zone furnace 10 4 . ~ ~ ~
This section includes a stopcock 106 and ball-joint 108 .~ ~:
arrangement used to achieve the low-pressure~ desired to carryout the reAction. This alternate sealing technique .:
requir~s lower temperatures for this portion of the set up, but allo~s. for rapid and nondestructive insertion of a glas~
"boat~ which holds the phosphorus and metal source~, ~he . ~`~
boat 112 (see Figure 15) also is designed to hold metal on `~
glass substrates 114 (Fi~ure 14) upon whioh the fil~ ~re to be deposited. These film/substrate configurations serve as initial starting points for device deqigns, as indicated below. - .
The section outside the furnace pr~vides a cold trap for vapor species. Specificall~,phosphorus, whîch is loaded into the zone s:losest to the out5ide sectiGn, is deposited .
.

-67- llO-OOlB

in the outside sectiO~ in large amounts, generally as the highly pyrophoric white form. Because this trap exists, the vzpor pressure conditions ~f the sys~em are quite di.~ferent from the totally-heated systems described above. It follows that the temperature conditions which successfully yield desired products in the first apparatus, are not appropriate for the second apparatus. The conditions appropriate for the latter were independently determined~

Example V
In the 54 cm long ~y 2.5 cm diameter quartz tube 100, with a 10 cm long by 1.0 cm diameter neck 11~, shown in Figure 11, phosphorus and potacsium were loaded, under dry nitrogen conditions, into opposite ends of the tuhe , in atom to atom ratio of 15 to 1. The potassium (99.95% pure) was loaded first ~y dropping small pieces, totaling 0.28g in wPight, into a cup 118 with khe tube oriented verticall.y.
The pieces were then melted and allowed to resolidify in the cup. The phosphorus (99~9999%~ was then added to the tube, the 3.33 grams of pieces easily being manipulated around the cup 118. The tube was then sealed by fusion of the neck 116, at 5 x 10 5 Torr.
The tube was then arranged in a Lindberg Model 54357-S
3-zone furnace to lie centered amongst the three zones. Un-like the Model 54357~ which has zone lengths of 6, 12 and 6 inches (15.2, 30.5, and 15.2 cm)~ the S model has zones of 8y 8 and 8 inches (20.3, 2003, and 20.3 cm). Two woven asbestos tapes, spiraled around the tub~, held it at the junotions of zones 1 and 2, and zones 2 and 3. ~ot only did these tape~- support the tube, they insulated the center zone from the higher temperatures of the outside zones. A
schematic representation of the resultant temperature profile is shown in Figure 1~7 A Honeywell Model DCP-7700 Digital contro'L Programmer was used to drive the three heat-ing zones through an appropriate warm-up period, to the 450 r 300, 450 gradient, which was held for 72 hours~ and then through a 15 hour cool down ~equence to ambient temperature.

-68- llO-OOlB

The materials formed in the tube were analyzed by the following procedure. First, in a dry nitrogen a-tmosphere, the tube was cut into seven -~ubular sec-tions, of approximately equal lengths, by use of a silicon carbide saw. Pieces of the ~ilms found in the sections (generally 10 microns or greater in thickness), were removed and individually examined by X-ray diffraction techniques. The remainder of each section was subjected to analysis by wet methods.
The P/K ratios oE the deposits Eound Eor the sec-tions are indicated in Figure 13. For the center regions, where T was approximately 300C, the bulk compositions were about 14/1, which falls within the accuracy limits of -the methods employed to identiEy the material as KP15. More revealing were the X ray powder diffraction patterns for the materials found having a P/K of about 1~, which clearly showed they matched those of KP15, either from single whiskers or bulk polycrystalline material. Furthermore, the pat-terns clearly showed the presence of both polycrystalline and amorphous materials in about a one to one ratio, as manifested by broadening of the peaks~
Example VI
The apparatus used in this example was modified relative to that of Example V. The quar-tz tube 119 was fabricated with "nozzles" 120 and 122 segregating the two end chambers from the cen~er one (see Figure 16). ~nder dry nitrogen conditions, melted potassium (0.~7 g, 99.95%
purity) was added to the outside chamber indicated at K, and allowed to resolidify. The addition tube 124 was then fused shut. Phosphorus (5.58 g, 99.9999% purity) was then added to the other outside chamber indicated at P and the whole apparatus evacuated and sealed at 1 X 10 5 Torr, by fusion of the second addition tube 126. The phosphorus to potassium ratio in the system was 15 atoms to 1 atom.
The sealed tube 119 was 41 cm long, and was centered amongst the three consecutive 2003 cm zones of a Lindberg -69~ 110-001B

Model 54357-S 3-zone ~urnace, T~:o thermal barriers (TB~ of woven asbestos tapes, spiraled around the tube, hel~ it at ~he junctions of zones ~ and 2, and zones 2 and 3~ In ad-dition to holding the tu~es, they insula~ed the center zone from the higher temperatures of the outside zones~ A Hon~y-well Model DCP 7700 Di~ital Control Programmer was used to drive the three heatin~ zones through a warm up period, to a 500, 355, 700C gradient. ~The phosphorus was at 500C, the potassium at 700C. The center zone tPmperature was select~
ed as 300C, but because the insulating characteristic~ of the woven tape are limited, heat spillover from the side chambers raised the oenter zone temperature to the 355C
le~el.~ This yradient was held for 80 hours, and then a 24 hour cool-down sequence was followed.
Whe~ tube 119 was cut open, under dry nitrogen condi-tions, using a silicon carbide saw, it was found that noæzle 122 between the potassium zone K and the cen~ex zone had become clogged with material, whioh looked like polyfibrous XP15. The center zone contained thin, light red films; : , thicker, darker red films; and several, r~latively large, ;~
monolithic ~oules. The two laxgest piec~s were each about 4 cm long, by 1 ~m wide, with a maximum thickness of about ~ , 4mm. One side of each pieee is relatively planar, while the other has a convex configuration, associated with growth against the inside walls of the circular reaction tube.
Wet nalysis of his material showed the potassium contant to be extremely low, as a bulk analysis 7 at less ~ ~ ' than 60 parts per- million. Electron Spectroscopy for Chemical Analysi5 ~ESCA) indicated that the potas~ium content of thi~ material decreased rapidly ou wardly of the tube wall on which it was first deposited, At 100 angstroms the ratio of P--K was about 50. As measured by ESCA the P K
ratio on th2 final surface deposited was in the or~er of 1000. X-ray d:iffraction studies showed the material to be amorphou3~

70- llO-OOlB

Example VII
U~lder dry nitrogen condit1ons, 0,l9g of melted potas- --sium t99.95~ purity) were transferred ~o vne of the outer-most sections 12B (5 cm long) ~ a pyrex boat 112 (Figure 15). The metal was ~llowed to resolidify. ~wo plain glass substrates 114 (see Figure 14), each about 7.5 cm long by 1 cm wide, were laid end to end, fillins the 15.3 cm long center section 130. Next, 1.36 grams of phosphorus (99.999%
purity obtained from Johnson Matthey) were added to the opposite outside section 132 of the boat. The phosphorus is i~ a mixed-size yranular form which readily pours out and fills in the bottom of section 132. Pyrex dividers 113 keep the P and ~ and su~strates from sliding in the boat 112~
The 35 cm l~ng boat 112 was then carefully slid into the 60 cm long by 2.5 cm diameter pyrex reaction chamber 134 of Figure 14, until section 128 with the po assium a~utted the round bottom, closed end o the ch~mber 136. A ~una-N :
O-ring, size 124 was then clamped into the O-ring joint 102, and the T eflon Stopcock 106 (supplied by ChemVac, Inc) screwed down tightly. On a vacuum line, the stopcock 106 was reopened and the chamber pumped down to 8 X 10 4 Torr.
The stopcock was then re-closed, sealing the reaction chamber.
The reaction chamber is arranged in a Lindberg Model ~4357-S 3-~one furnac:e~ As shown in Figure 14, two woven-glass tapes 137 and 139, spiraled around the tube, supported the chamber at the junctions of zor~es 1 and 2, and zones 2 .
and 3. The~e tapes forming thenmal barriers (TB) were set to just li ~ompletely within the center zvne. A third ~.
spiraled tape 138 was u5ed to ~upport and thermally insulate ths point where the apparatus exits the heating chamber of the fur~ace. A cylindrical plug 140 o~ a ceræmic like material was used to stem heat loss out of the furnace opening at the other end of the cbam~er.
This arrangement of the apparatus re5ults in section 128 of tbe boa1 112 containin~ the p~tassium to lie within the third heating zone, section 130 containing sub~trates to lie in the center, or second, heating zone and section 132 of the boat containirl~ p~osphorus to lie in th2 first heat~

-71- 110 OOlB

ing zone. I-t also results in a large segment of the apparatus being outside -the furnace, at ambient temperature.
A Honeywell Model DCP 7700 Digital Control Programmer was used to drive the three heating sections through a war~-up period in which the temperatures were brought to 100, 150, 100C in the phosphorus zone, the substrate zone, and the po-tassium zone, respectively. Then, as rapidly as possible (approximately 18 minutes) the gradient was driven to 500, 300, 400C, where it was held for about 8 hours. The furnace was then allowed to cool at its inherent rate, to a profile o-f 100, 100, 100C, which took about 10 hours. The furnace then was allowed to cool to room temperature.
The tube 134 was removed from the furnace. The section ou-tside the furnace contained deposits of white, yellow, and yellow-red materials, all of which were probably phosphorus in varying stages of polymerizationO The phosphorus heating zone was clear of material, while the potassium zone contained a variety of materials, ranging in color -from tan, to yellow, to orange.
The latter extended slightly into the center ~one, which otherwise was covered through one-halE of its length, next to the potassium zone, with a dark film, which transmitted red light when a source lamp was shone through it. The remaining half of the zone was clear of material.
The apparatus was opened under dry nitrogen conditions, the pyrex boat 112 withdrawn, and the glass substrates, covered with the red film, removed from the boat, and placed in a tightly sealed bottle, for later analysis. (When the remainder of the materials were exposed to ambient conditions, the phosphorus deposits in the exposed section of tube would generally burn vigorously, though those closest to the phosphorus source did not exhibit such reactivity. The materials which were in the potassium-source section of the apparatus were very reactive when exposed to moisture. They generally burned vigorously, apparently by the production of hydrogen via reduction of water).
The technique was repeated several times. Further examples are noted in Table XII.

-72- ~LZ~ llO-OOlB

z; ~ ~ m E~
H :r:
O W ~ U- o o o C: o U~ o :Z U ~ 1 o o o U~ U~ o o u~
G t~
E~ ~ Z o z o ~ u~ o ~ o E~ ~ O ~ o ~n ,,~ o o ~J ~' ~ ~~'` ~ '~
~.r~ Z O ~ O
U~ Z ~ f:l ~
o o o o H U:l H ~ r~r~ ~ N
X ~. ~ ~ ~
W ~) t-l r~ ~ ~ r~l ~ ~1 ~1 1~1 m U~
~: ~~ ~ ~ U~ O ~ ~ O
.. .... ~. .
~ C~ O O O OO O O G

W OO O O O C:~ O O O
~ U~
E~ m U ~o O U~ O OO O O
u~ K o O o o ~~ O e~o o o D O~
c~o o o o o o o o ~ ~ O o o o c~ o ~ ~
P~ Z r~~ ~ ~ t~ ~1 ~ ~ ~1 C~
~ V o o o m o oLn Lr~ o E~ ~ o oo o ~ o u~ ~ ~ o W O ~D t` 00 ~ O
~:; Z ~1' ~ ~r ~ u~ U~ u) ~ U~

-73-~ ~ 110-OOlB

There exist limiting condi-tions for the prepara-tioQ
of the dark films which transmit red light. If the temperatures in the two source zones are dropped slightly, as in run number 49 o-E Table XII, the amount of material formed, as manifes-ted by the length of the deposit, drops dramatically. Similarly, subtle differences between the performance characteris~ics of two otherwise identical Model 54357S 3-zone furnaces require that in the second furnace (B), the temperature of ~he phosphorus source be raised to higher temperature (see run numbers 50, 51 and 52). Raising the phosphorus source temperature to 550C gives a good result, raising it to 525C gives a better result.
Analysis of materials from runs 46, ~7 and 48, by Scanning Electron Mlcroscope with electron diEfrac-tion analysis (SEM-EDAX) me-thodologies revealed the material to be films, on the order of 6-7 microns in thickness, and to be of an amorphous character, with no discernible structure evident in the micrographs.
Summary of Vapor Transport Conditions Processing features for controlling product types are: 1) Use of a three zone furnace for more uniform temperature con~rol; 2) Extended tube length; 3) Use o-f thermal barriers for temperature gradient control; 4) Use of thermal plugs at ends of oven; and 5) Use of extended narrow addition tube to obtain cylindrical boules.
Ranges of conditions for one source vapor transport are:
1) Reaction zone temperatures range from 650-550C;
Cold zone deposition temperatures range from 450-300C.
2) Deposition temperature for single crystals of KP15 were found to range plus and minus 25C around a center val~le of 465-475C.
3) Deposition temperature for polycrystalline -films were found to range from about 455C down to 375C

-74- llO-OOlB

4) Deposition temperature for amorphous forms of the new form of phosphorus range from about 375C down to at least 300C. (~Jo lower temperatures were investigated to date).
The conditions for two source vapor transport for forming bulk KP15 materials are (Figure 11 apparatus);
Phosphorus, temperature at ~50C, Potassium at 450C, and deposit zone at 300C; deposits were thick films of mixed polycrystalline and amorphous KP15; for bulk amorphous KPX (x much greater than 15 the new form of phosphorus, Figure 16 apparatus): Phosphorus at 500C, Potassium at 700C and deposit zone at 355C. K source became plugged, deposit was bulk amorphous KPX; for thin films of amorphous KP15 (Figure 14 apparatus) Phosphorus at 500C, 15 Potassium at 400C, and substrate at 300C.
For thin films of KP15, the Phosphorus source may be raised to 525C and amorphous KP15 is still produced.
If the Phosphorus source temperature is dropped to 475C, the system does not yield KP15. If the Potassium source temperature is dropped to 375C, the system does not yield KP15. The substrate temperatures may be raised to 315C
and the system will still yield KP15, but not if they are raised to 325C.
Preparation of Polycrystalline Metal Polyphosphides in Large Amounts Via "Condensed Phase Synthesis".
Although not formed in a physical state appropriate to the tapping of their useful semiconducting properties, alkali-metal polyphosphides of the type MP15, MP7, and MPll can readily be prepared in gram or more quantities by a technique we call "condensed-phase" synthesis. Before using this technique, the reactants are generally brought in intimate contact by a ball-milling procedure. Decagram or more quantities of the elements are loaded in ball-mills, under dry nitrogen conditions, in the desired metal to phosphorus, atom to atom ratio, e.gO P/M 15 to 1 for MP15. The sealed mills are then utilized for 40 or more hours to reduce the -75~ llO~OOlB

components to a well-mixed, homogeneous, free-flowin~
powder. The mills are generally heated during 20 hours or so of -the milling, to about 100C. This is done to increase the fluidity of metal component during the milling.
A portion of the milled mixture, ~enerally 10 grams or more, is transferred to a quartz ampoule, under dry nitrogen conditions. The ampoule ranges in size from 2.5 cm in diameter by 6.5 cm in length, to ~.5 cm in diameter by 25 cm in length, depending on the charge si~e to be processed.
The tube is sealed at reduced pressure (generally less than Torr3.
The reaction is carried out by subjecting the tube to an ever increasing temperature, under isothermal conditions, until the applied temperature reaches 500 or 525C. By isothermal conditions we mean that the whole mass of material is always as nearly as practicable at the same temperature to prevent vapor transport -from hot to cold portions which would result in non-uniform products. The highest soaking temperature is held for a substantial time, during which a powdery polycrystalline or crystalline product is formed. A typical soaking time is 72 hours. The longer the reaction, or soaking time, the more crystalline the product ~as maniEested by grain si7e, sharpness of X-ray powder-diffraction lines, e-tc~. The hot tube is also taken through a cooling period (more than 10 hours) to ambient temperature, Slow cooling is not necessary for the reaction, but prevents tube breakage due to -the different thermal coefficients of the products and the quartæ ampouleu Both the heat-up and cool-down periods have been observed to best be devised as relatively long (more than 10 hours) with soaking at intermediate temperatures (e.g., 200, 300, 400, 450C.) for 4-6 hours. Failure to follow these slow heat-ups or cool-downs often resulted in explosions of the reaction tubes. However, the products of the condensed phase reactions were the same as in slow cool down except that a small quantity of residual phosphorus would be white rather than recl phosphorus.

-76~ llO-OOlB

Example VIII
19.5 grams of a ball milled mixture of reagent grade phosphorus and potassium, in an atom to atom ratio of 15 to 1, was transferred into a 6.5 cm long by 2.5 cm diameter quartz tube, which tapered to a 8 cm long by 1.0 cm diameter section. The transfer was carried out under dry nitrogen conditions. The tube was sealed at reduced pressure (1 x lO 4 Torr) by fusing the narrow section a centimeter or so above the wider part of the tube.
The tube was supported in the center zone of a Lindberg Model 54357 ~hree-zon~e furnace by a second quartz tube, or liner, which was, in turn, supported in the radial center of the heating chamber by asbesto~ blocks. The 3-zone furnace heating elements were driven by a Honeywell Model DCP-7700 Digital Control Programmer which enables processing to be preprogrammed and carried out in a reproducible fashion. Using the programmer, the reaction tube was subjected to the following temperatures for the indicated lengths of time: 100C, 1 hr; 450C, 6 hrs.;
500C, 18 hrs.; 525C, 72 hrs.; 300C, 2 hrs.; and 200C, 4 hrs. ~When all three zones are controlled at the same temperature, the center ~one is highly isothermal, with a temperature variance of less than l~C across the 7one).
After the furnace cooled to ambient temperature, at the inherent cooling-rate of the furnace, the reaction tube was removed from the furnace. Under dry nitrogen conditions, the quartz ampoule was cut open using a silicon-carbide saw, and the dark purple, polycrystalline mass removed. A sample of the material was subject~d to compositional analysis. Wet analysis gave a P/K ratio of about 14.2 to l, which is accurate to about 6% of the theoretical valu~ of 15 to 1. Products from similar runs on K/P15 charges fell range values, as shown in Table XIII.

~77~ o-oo~ 13 :~ ~ O O O O O O O O O O O O O O O O O O O O O O O O
X X ~ X X X :~ X ~ X X X X X X X X X X X X X X X X
~Z~ ooo<~-ooo~ooo~oooooooooooo~r O E' a: ~ rl ~ 1~ I N _~ t'N~ ~ 5~ N ~ 1 N 1`1 .-~ ~ ~) N N _l ~E~ ,noooooooooooooooooooooooo o u~ N N U~ O O ~ D V
E ~ 3 X --I ~ N N N _I N N ~ ~
~, Vl 5 E ~J O N N N N N ~ N N O O ~ O U'l o O O O 0 N ~ U~
.' C~ 3:
,, C:
~ ~ W
X C U) N N m N 1~ t~ ~ ~ U ~ ~ N ~ 0~ U~
1:1 W O U~ ~ N ~ r~ I m ~ N

o tl E~
C~ ~n o N ~ o ~n m m o u~ o . ~ . ~D
o L) ~ l~ l` N ~
O~ ~ X ~ o~ z o~ z ~ z z z z ~ ~

~n ~ o N ~ o~ u~ a N o~ a Q ~ Q n L'~ L ~D ~r ~r ~ ~ ~ ~ u~ z o~ z ~o Z z Z Z Z Z

~:~ Y~Y~Y~Y~Y~x~x~ zz~ù~zY~Y~YYx~x W O L'~ O _ N ~ ~ L~ ~O ~
;~ L~ L~l Ul U~D V ~ ~D ~D ~ ~ I~ r- r 1~ r~ 1~ r~ 1~ r~ 1~

-7~- llO-OOlB

In addition, several samples from di-ferent runs were subjected to mor~hologlcal analysis. The XRD powder diffraction patterns for these materials were readily matched to those obtained from the single crystal KP15 samples produced by the vapor-transport me-thods cited elsewhere.
The methodology was carried over to other ~etal-phosphorus systems, as is indicated in ~he -~able.
Comparisons of the XRD data of these materials, both with each other and that obtained on single crystals established the analogous nature oE the products, i.e. they all have basically the same all parallel pentagonal tubes of covalently bonded phosphorus.
Milling Metals with Red Phosphorus Introduction We have utilized ball milling to prepare homogeneous, intimately contacted mixtures of red phosphorus with Group la and group 5a metals.
The uilled produc-ts are relatively air stable and they provide conveniently handled starting materials ~or the previously described condensed phase and single source vapor transport techniques. Their stability indicates that poly-phosphides have formed at least in part during the millingprocess.
Summary The Group la metals (with the excep~ion of lithium) have proved to ball mill easily with red phosphorus. The facility of milling becomes even more pronounced with the lower melting metals, typified by rubidium and cesium. A
problem arises when the Group la M/P ratio is varied from 1/15 down to 1/7. The increased metal content generally results in severe agglomeration of the charge onto the walls of the ball mill. Fortunately, the agglomerated products are easily scraped from the mill and crushed through a 12 mesh sieve. Lithium and arsenic are somewhat di~icul-t to ~79~ 110-OOlB

mill using ~he standard b~ll milli.ng procedure due to their hardness and higher meltin~ points.

Reagent_Purity The initial experimental work used reagent grade metals and reagent grade phosphorus. We now use only high purity metals and electronic grade (99.999% and 99.9999% pure) red phosphorus obtained from Johnson Matthey.

Mode of Millin~
A. Standard Ball Millin~ (Rotation?
This was originally the method of choice for the alkali M/P systems. However, we have used more intensive grinding processes (cryogenic and vibratory milling) for the other group 5a metals.
The stainless steel ball mills were fabricated "in house" and as shown in Figure 17 compriæe a cylinder 150 with these dimensions --4.5" O.D. ~ 6" height X 1/4" wall thickness~ The ~op of the mill i5 provided with an inner flange 151 to accept a Viton O-ring 152. A stainless steel top 154 is held in place by a bar 155 tightened down with a screw 156.
One mill has smooth inside walls. The second mill was constructed with three baffles welded onto the walls from top to bottom. These act a5 lifters for the balls and re-agents and result in more efficient grinding.
A total of less than 50-60 g reagent charge is desir-able. Initial milling experiments used 1/4" stainlPss steel balls; we have since achieved better results with a mixture of 1/4" and 1/8" stainless steel balls.

Cryogenic Milling ~-196C~
Tnis was accomplished usin~ the Spex freezer mill (available from Spex Industries, Metuch~n, N.J.).
Due to equipment limitations, only small quantities (2-3 g~ c~n be milled in a single operation--however, this can be done quickly at liqui~ nitrogen temperatures (in a *Trademark .

-80- ~ X ~ 110-001B

matter of a ew minutes~. Thus, this technique finds appli-cability in xeducing to powder 'orm, the harder and higher melting metals such Z5 lit~ m and arsenic~ These can then be co~ground with red phosphorus in the rotating ball mill or vibratory mill.

Vibratory Milling The equipment (Vibratom~ is available from TEMA, Inc., Cincinnatil Ohlo.
This is essentially a ball mill, but instead of using a rotating motion, circulax vibrations are generated--similar to that of a paint shaker. The dimensions of the mill are 5 1/4" O.D. X 3.5" height X 1/8" wall thickness.
The mill does not contain baffles. We have used this mill for the difficult to mill elements such as Aso ~ime of Milling There has been considerable variation here. Generally, the duration of hot milling is not less than 40 hrs. nor more than 100 hrs. To some extent, this has been determi~ed by the system being milled. Le~s time is required for the lower melting Cs and Rb systems.

Temperature of Millin~
This has either been at amhient temperature or the mills have been externally heated to approximately 100C
with a heat lamp. Ambient temperatures are suitable for 7OW
melting point meta~s such as Cs (28.7~C) and Rb ~38.9C), External heat lamp application to 75-100C for 3-4 hours was definitely beneficial for the Na (97.8C) and K (63.7~C) systems. Heating to 100C was of no value with Li ~108.5C). We conclude that stable products are the result of milling melted alkali metal and phosphorus.

z~

Ball Milling of K/P15 Example IX ~Reference No. 88, Table XIV) Under nitrogen in a dry box, an un~affled stainless steel ball mill containing 884 g of 1/~" s~ainless steel ball~ was charged with 6.14 g (.1S7 atom ) 99.95% pure K
(from United Mineral and Chem. Co.) and 72.95 ~ (2.36 atom) of 99.9999~ pure red P ~from Johnson Matthey Chemicals).
The mill was s~aled and rotated on a roll station for a total of 71 hours. The mill was heated to approximately :- :
100C for 4 hours by playing a heat lamp on its surface.
The mill contents were dis~harged in the dry box to a 12 mesh ~ieve and pan. No agglomeration of the product was obs~rved. The steel balls were separated from the product on the sieve. A total of 76.~ g of black powder product -i was obtained.

Ball Milling of Cs/P7 ~, , ., . .: ~.
E~ æ~ (R~ference No. 115, Table XIV) ;~
Under nitrogen in a dry box, a baffled stainless steel . ~-ball mill containing 450 g 1/4~ and 450 g 1/8:' stainles~
ste~l ~alls was charged with 1~.12 g ~.091~ atom~ of 99.98%
pure C5 Srom ~l~a/Ventron Corp.) and 19.77 g ~.638 atom) of - `~
gg.g99~ pure red ~ (from Johnson Matthey Chemicals), The mill was sealed and rotated on a roll station for 46.5 hours at ambient temperature. (no extern~l heat source appli~d). . ~ , Upon opening the mill in the dry box, a}most total agglomer- ~ , ation of the pr9duct was observed on the mill wallsD This material was scraped off with a spatula and discharged to a 1~ mesh sieve and pan. Th~ chunks of product were then crushed through the sie~e; A total of 27.8 g of product was collec~ed in the. pan.
Tab~e XIV summarizes the results of milling various ~ ~
m2tals with re~ phosphorus, As previously noted, the~e.~ j materials are s~rprisin~ly stable~

Z~ 110--OOlB
.C) b~
r O ~ O ~ D O ~a O
~ . r~ , ~ ~ ~ ~ ~ ~ E ~
g~ r~ O ~ k r~ ~ r).~ ~ r~ r~ ~ k rl~ 3 ` ~ ~ ~Q E 5~ Q Q ~ D o Q Q
e ~
v, ~ v~

~J V '~ r w ~ ~
~ c) ~ a ~ u v u u u r~ n ~ n a v ~ v O O O v P e ~ O ~ ~ O

O O O 0~ 0 ~ Ul O O ~ U) C~
S ~ Ul ~ O~
o ~ ~ a~ ~ ~ a ~ ~ ~ ~ h ~ ~ 1 v u n ~ r~
~ X r2 ~ Q r~ m m m m ~ r~ Q m ~ ro ~ m ~ ~ ~ O ~ r~
x 2 ~p~ ~ v r 8~ e 6 ¢ v a ~ a c a Q ~D CJ Q t Q tl Q ~1)~0 6) ID ff' :~ k ~ k k 1 1 i - k 1 ~ h ~ I I k O E~ ~ v V v V V V V V C V ~ V 0~ ~ ~ V C

),J h h )~ h C~ k O~ I ~ I h IC D. X P~ X ~ X tl. X P. K t'. K D. ~: r~ X ~Ç :C P. X
V ~ u~ o co o~ o o 0 0 o o ~3 W . u~ 1 . O . O . O . ~ . C~ . O N . N ~ o~
5 r~ O ~ 1 t~ r ~ ~ r~ ~ w r~ u) L l _ cr u~
l O N O U'l U- ~ Ul U'l Ul US i~
U ~~ ~.e X X X X X ~ ~ ~ ~ X ~
~4 O Z _ "., . 1 ~ IU Ql V ~ O ~ -- ~
O ~1 al la d 0 ~ 'J ~ d ~) U .a E E ~ m ~ E x r4 r O co m so ~ o ,_~

~3 I10-O~
a~ C ~ _4 .c a~--~ g z F.
4~ o ~- c~ ~- cJ E
r~ ~ ~ v t~ v o~ o ~
~o ~ ~1 0 _ ~o ~ > 0 S ~ O .C ~ g 1 c ~ c~ ~ ~ v v'~ c~
r lo o n IP h ~4 g o 0 0 _I E
~ ''~ r ~ ~'~ ~ ~'~ E ~ G' C- _1 # _ ::~ N ~ > I` U~ ~ E ~
In ~ 1`1 N o al ~ ~ U~ /J' r~ U~ U~ O
~: ~ 0 1 ~ ~ ~ C ~
. --_ _ _ _ ~_ ,C h V
::~ ,C.C .C .C ~ U~ .
r~l 1''1 ~ 1~1 1' u~ N 1l _I o V
D. D t~ E ~
:~: o o o o O O O ~E ~ J
~ û~ S ~
~ z ~ O O ~ ~ O ~ ~ ~ ~ ~ X-- a ~ E
U~ O :1 o~ ~ co 0 N ~ 0 .IJ V ~ 2 ~ ~ ~
~ E ~! ~ ru~ ~D N _~ h ~ ~ h I O

¦ a h r ~ h hul h 0 ~ V c~ cr v cr v :E s H ~ E ~ X X.Y X X ~ 0 C U ~ 0 0 U ~ O 1: 1 z -- ~ 3 m alal m ~ o E ~. c~ h ~ l ~ O O
U E~ ~ E~ U 'U o E~ O ~ E~ .t t) O
? ~n u~ 1~ r ~ ~ ¢
x .a a~ ~ ~ c i ~ ~ '~ L'l V ~ ,~
I W ~ 0 ~: S r t~ rU~ b~
O v ~ r o~ ~I X ~ J~
V co tP S ID C ~ c~ a- c~ ~ C) o, ~ L'~ E
Z ~: ~ ~ a 00 ,4 . V~ U .
a ~ c ~ r~ r~ ~o I o~ ~ U 1~; ~ U o~

~S .. ~, ._~ . ~ .u~ ~n V-~ o ul . ~ . . U~ . u ~
D: U~ n U~ t~lCD 17 N U~ N ~ 1~ N

~ r~ O r-I` I` i` Y~ 10 Ul ~-- :C K X ~ X x :.:
~ ~ s. c~
Z ~ ro co '~ r Q ~1 U ~.a .a ~ so ~_ ~: ~ ~ E E ~ V

.
r~l ~ ~ u~ ~D

~5 ~
-84- llo-Ool~

,C ~, ~. o E :~` t,~ 0 t> ~ o O.C g V` Cl O V V O C) O ~ O C O
o ~ c~ o ~ t o ~ o o o a o n ;~ o V~ 1 h1,1 ~ Ql 0~ t` 1~ g 0, rJ
~ 8 ~ a 8 .Y o o o 8 ~ ~D ~
~1 r ~ ~ t ~r r U ~D r1'') 1~. U- ~ o~ r E
~: O ~ ~ 1 U r l ~ m -W ~ o b JC ~ l m h x m X-- t: ~-x.t: x Ul Ic X C
W ~C ILI U OO W t Ul P ~r~ E~ ~`Dtl.~ t~ U~

C ~ ~ ' ~ o o U c~~n o ::~ ~ ~d ~ ~u~ D ~ O r~ c~
C ~ :E: ~ ~ ~r_ CD O 1` ~D 1`
E O
O CC C ~1IC r .~ IC ~ ~ t IC
aD. oo o v tl O ~ V O ~ O O
~~ h ~h ~ cn i ~ C t~ U~ h h WtL~ t ~ Wo 0 110 ID C) C C IC 0 C X .~C IC E ~ I t. ,!~t t_ O rl (~
E~ _ t~ ~ ~ ~. m ~o m ~ ~n 8 ~ E ~ 0 E ~ t~ t_ O E~ ", ~ S E~ 0 ~ L ~ ~ O e ~ ~ ~ t ~ o ~ V~ o s ~ E~ ~ 8 s 3 ~ u~ ~ ~ "~ o~ ~ I C Q ~ ,e ~ IC C ~
X ~ J
W E ~ r o~ C U
~ t~ ).1 0~ b 111 115 L~
C ~; ~
~4 ~ ~ ~~ O~ V
O :~C ~C ~ v~ ~ ~ a, Cl`~ ~ ~ ~ U~ OX ~ ~ O U~ Cl t Z ~ O ~ ~ t~ t~ -~ 0~ t~ ~ t~C~ ~ t~ . tl~
I O ~ ~ h O~ b ~ I o 7 K ~ t.~ 4 z 14 z ~b I t4 I t t ~ N ~ a~ I~ N O~ N . ~ . D
hJ ~ 1 N r ~ N
r~
!n ~3 _. t ~ ~ "~
~0 ~ t , ~r~l ~ ~-1 p,~ ~ D.P- ~'~ 0.
~3 o z - - - ~
:E: ~ 0 t~

. .
~! r l _I C' D o z o~

-85- lltl-ODlT.~

~ o ~ h D
1.~ S~ C) h L'~ Et7` E ~ Sn G E ~ ~ E tP v 6 c~ v, G
::1 ~1 N E V N E 1~ I U) E ~ Co E~ o ~ ~o ~ ~ v D ~ In V ~ ~ ~1 so N _ I;) N ~

V V c V c V V V

t~ ~
~ZC'.
u~ ~ ~ _ o u~ o o ~n s~:E o~

8 l ~ ~ v = ~ sh I v o v O v O v o _ ~ ~4 :~ r :~ o ~ 1 ~; o ~ a O > r ~ e :- v~ ~ lo n~ ~ ¢ ~
x ~c ~o ~ ~: ~
E~ ~ v ,~, ,,, SJ' o c~ 1~ cP ~ o~ ~o cr~ o o~ ~ ~ ~ o~ s~ c, C~ ~ ~ s ~ ~ s c~ c~
Z Z ~ ~ s~ cr-~ ~ ~D N ~r ~ ~ ~ ~ l ~ ~ o r~
W ~ . o . ~ r O O r~ s~

~ O E s" u~
~ V U V ' ~ ~ ~
_~
O Z ~ _ ~ ~o ra ~ ~ C s~
~ ~ ~ ~ n r A D
1 ~ :E:
r~ O q _ s ~D O ~ ~ ~

, -86- llO-OOlB

Analysis of Products Table XV summarizes the various MPX (X = 15 and x much greater than 15, the new form of phosphorus) materials synthesized from vapor transpc~rt with one source (lS-VT), from vapor transport with two source (2S-VT), condensed phase processes and chemical vapor cleposition ~CVD).
TABLE XV
MPX M = Li, ~a, K, Rb, Cs X ~C = 15 X much ~reater than 15_ single X X
lS-VT crystals poly. B, TF
amorphous B
single 2S-VT crystals poly. TF TF
amorphous TF B _ _ single Condensed crystals X
Phase poly. B*

C V D amorphou~ TF _ _ _ _ _ X~ = crystals/whiskers B: = Bulk greater than 10 micrometers thick TF: = Thin film less than 10 micrometers thick B* = Powder 12 L~
-87- llO-OOlB

The materials obtained from these techniques were crystals or whiskers, referred .o as X; solid polycrystal-line bulk, referxed to as B; solid thin film, referred to as TF; solid amorphous, referred ~o as B and TF; and bulk powder from condensed phase synthesis referred to as B*.
The analysis of MP15 crystalline materials was given above with reference to Figures 7-10. As indicated in Table XV, the polycrystalline and amorphous MP15 materials have only been produced i~ the form of thin films.
Polycrystalline bulk and thin films of KPX (x much greater than 15~ were obtained by vapor transport (one source and two ~ources). These polycrystalline thin films nucleate on glas~ substrates (or glass walls) and show dense packing o parallel whi~kers growing perpendicular to the substrate. SEM photomicrographs, Figures 18, 19, a~d 20 of such matexials show a large physical separation between the KPX whiskers.
These polycrystalline thin films arQ formed at low temperatures fro~ around 455C to 375C wher~ the amorphous phase begins to form.
Analysis on these materials wet chemical, XRD and EDAX
consistently show x to ~e much grea er than 15 (typically greater than 10003. A typical powder XRD diagram f}ngerprint of crystalline MPX ~x much greater than lS) is shown in Figure 10.
As indicated in Table XY, ~morphous MPX materials can he formed in bulk form lboule~ by the vapor transport tech-niques. Thefie boule~ are fonmed in the narrow end 160 of ~ube 32 ~Figures 1 and 2), the narrow end 162 of tube 5~ o~
Figure 3p or as pieces of material in zone 2 of Figure 16.
~hese materials show no X-ray diffraction peaksD
XRD powder diagrams were used in our study to charac-terize the degree of amorphicity of the materials obtained ~Lf~ 5;~L
-88- llO-OOlB

by these tcchniques. These amorphous MPX materials where x is much greater than 15 can be cut, lapped and polished using conventional semic~ductor techniques for wafer processing. This is even true of material containing no more than 50 to 500 parts per m:illion of M, a new form o~
phosphorus.
The resulting high x, KPX amorphous wafers or substra~es were shown to have useful semiconductor properties with electro-optieal response almost identical with whis~ers of KP15. W~ therefore conclude that the local order of all MPX materials where x = 15 or is very much greater than 15 (when solidified in the presence of alkali metal) exhibit the same local order substantially throughout their extent. This local order is the all parallel pentagonal phosphorus tubes.
Amorphous high x, XPX material~ were pr~pared with mirror finish surfaces for electro-optical evaluation.
Routine suxface preparation of these amorphous materials includes several processing steps such ag cutting, embedding, lapping,polishing, and chemical etching~ Surface work damage induced during such processing steps are known to af~ect the electro-optical performance of semiconductor materials. Therefore, attention was focused on assessing techniques ~nd processing steps leading to a "damage free surface. The following processing steps have been found to be suitable f or the preparation of high quality mirror finish surfaces.
Embedded boules of high x, XP~ (about 1 to 2 cm in lengt~) from Table VII w~re cut with a slow speed diamond saw using minimum press~re. Each wafer was sliced to a thickness of appxoximately 1 mm~ The wafer was then immersed in a bromineJHN03 solution. To remove sufficient cutting damage the thickness of each wafer wa~ reduced by this chemical letching by approximately 50 micrometers. The wafers wer~ then washed and ch~cked for inclusions and voids. The h:igh x, XP~ amorphous material app~ars to be void freeO

-89- 3L~ 2~ llO-OOlB

A standard low temperature wax (melting point about 80C) was used to mount tha high x, KPX wafers onto a polishing block. The wafers were then lapped at 50 rpm at 2 minute intervals individually with a 400 and 600 SiC grit using distilled water as a lubricant with a 50 g/cm2 weight until a smooth surface was achieved.
The final polishing step was carried out for one hour at 50 rpm with 50 g/cm2 weight on a Texmet cloth with 3 micrometers diamond compouncl and lapping oil as extender.
This polishing step was followed by an additional fifteen minute polishing step at 50 rpm with 50 g/cm2 weight on a microcloth with a slurry of 0.05 micrometers of gamma alumina suspension in distilled water. All procedures require scrupulous in between cleaning steps in a sonic bath with subsequent rinsing and drying.
Samples prepared by this technique have a high quality mirror finish surface. The final polishing step was performed on standard metallographic Buehler polishing equipment.
Chemical etching plays a prominent role in wafer preparation, sur~ace treatment, pre-device preparation, metallization and device processing.
Numerous review articles are available covering the chemistry and the practical aspects of etching processes.
However, most information on specific etchants is widely scattered throughout the scientific literature. ~n attempt was made to bring together essential information that should be useful to the selection of an etching process relevant to these amorphous high x materials, Special att~ntion was placed on etching procedures and processes used for surface preparation. It was found that some of the etching solutions and procedures currently used for GaP and InP are applicable but with difEerent etching ratee.
The following etching solutions were selected and tested:

-90 llO-OOlB

- 5-10% Br2 95~90% CH30H for general etching and polishing - 1% Br2, 99~ CH30H for polishing high quality surface (approximately 1 micron/minute) - 5~ by weight NaOCl solution for chemical polishing - 1 HCl : 2 H~03 (1% Br2) for removing work damage after cutting and lapping - 1 HCl : 2 HN03 for removing surface layer.
Several samples were prepared for optical absorption measurements. The above technique was used to slice and polish on both sides amorphous wafers of high x material as thin as 0.5mm. Reference samples of GaP and GaAs crystals were also polished on both sides and used to measure the band gap by optical absorption.
Etching techniques were developed to reveal microstructures and to thin down small areas to 0.2mm thick for optical absorption.
Several etching solutions were selected and tested. The best chemical solution was found to be a mixture of 6.0 g potassium hydroxide, 4 g red potassium ferric cyanide and 50 ml distilled water at 70C.
Application to reveal an etching pattern takes less than 60 seconds. This solution is v ry stable and can be used with reproducible etching rates.
After embedding, cutting and polishing, several samples of amorphous KPX, (x much greater than 15) from Tables VII and VIII have been etched. Typical microstructures were revealed from this chemical etching treatment after 30 secondsD
Figure 21 is a photomicrograph at 360 magnification of the etching pattsrn on a surface cut perpendicular to the axis of an amorphous boule o~ high x material grown by single source vapor transport ~Ref2rence ~o. 28, Tabla VII) showing honeycomb microstructures with well defined domains a few microns in size. These honeycomb microstructures are characterist:ic for an etching pattern on a material having a two dimensional atomic framework (such as parallel tubes~.

-91~ llO-OOlB

Figure 22 is a photomicrograph at 360 magnification of the etching pattern on a surface cut perpendicular to the axis of growth of the amorphous high x material grown by two source vapor transport in Example VI~ ~igure 23 is a photomicrograph of the same etched surface as shown in Figure 22 at 720 magnification. Figure 24 is a photomicrograph at 360 magnification of an etched surface perpendicular to the surface shown in Figures 2~ and 23 and shows an etching pattern characteristic of tubular packing.
Thus we conclude from the available evidence that our MPxmaterials where M is an alkali metal where x is much greater than 15, i.e. where the amount of alkali metal is as little as 50 parts per million,all have as their local order the pentagonal phosphorus tubes aither all parallel (the MP15 form) or double alternating perpendicular layer (~onoclinic phosphorus).

Electro-Optical Characterization of High Phosphorus Materials from One Source Vapor Transport The electro-optical characterization was carried out on single crystal whiskers, on polycrystalline films, and amorphous films and boules. The characterization consists o~ (1) optical measurements on samples with no electrical contacts (absorption edget photoluminescence) ~2) electrical measurements with simple contacts o~ linear behavior (conductivity, temperature dependent conductivity, photoconductivity, wavelength dependence of photoconductiYity, conductivity type) 13) electrical measurements with non-linear or rectifying con acts with metals which are indicative of the semiconducting behavior~
From the above data we extracted the properties which indicate that all materials produced have electrical criteria for useful semiconductors, that is, they all have an energy band gap from 1-3 eV; conductivity between 35 10 5-10 12 (ohm-cm) 1; a photoconductivity ratio from 100 to 10,000, and chemical and physical stability under ambient operating conditions.

-92- llO-OOlB

Measurements were carxied out on the following equip-ment:
~1) Absorption ed~e - Zeiss 2 beam IX and visible spectrometer Photoluminescence - Low t~perature (4K~ cryostat and laser exci~atiOIl ~2~ Conducti~ity - ~ probe and 4 probe measureme~ts Temperature Depen- ::
dent conductivity - from 300K to 550~R in an evacu~ ~;
ated ch~mber Photoconductivi~y - with light source of 2 : .
approximately lOOmW/cm Wavelength dependent photoconductivity - Xe lamp light source and mono~
chromator Condurti~ity type ~ thermoelectric power measurement with hot and cold probes (3) wet silver paint was used to provide a temporary ~z:~
junction to materials~ ~-ith a photovoltaic open circuit ~oltage of G.2V measured under illuminationO
Metallic and pressur~ contac s forming junctions were evaluated for their current voltage characteristics on a Tektronix curve tracer. ~ ¦
Data on samples from the broad class of materials under ! ,''~
investigatio~ are summarized in Tables XVI, XVII, XVIII and XIX~
Table XVI summarizes the basic physical, chemical and electro optical properties of the prototype matPrial namely, ~ -KPX, x ranging from 15 to much greater than 15, in various physical forms and chemical composition. :

-93- llO-OOlB

C) ,~
a .,1 X~
P~ a X~
-- ~
h :.
Lr~ O 00 s P. ~
t I ~1 1 O ~
~ O O
8 ~ x ~ , , ~ , I
h n ,_~
X
O ~ ~

h Z; h H al o ~ a E~ s~
tn ~ I Q) C~ C~

H ~ O O O
~ o ~ x ~ ~ ~ ~ ~ ~ ~ r.o m ~ ~
E~ .~ _I
~ K
R
~, ~:
E~ O
C~ ~ o , I N ~ ~ ~
o o O ~
:Z
.,~
u~
o a) ~1 1 0 o ~ ~ ~ r. , 1 ~ ~ ~ ~ W ~
I . o ~ (3 0 a) ~ o t3 ~ h U~ Q
1~1 O ~ a o ~~ o o) . ~ o u a!
H l,~
E~ oIJ I O ~ O ~ m o 1:
d Fr h o ~ ;5 0 O ~ O ,Y ~ 0 ~2~ 0 p,. ~
~rl P O ~O O ~ - I~ ~a rl 1~) O ~ rl ~; r! O ~ ~ O O
Q. O ~ ~ ~ o ~ o ~ O E~ O E~
~ ~;,~ o .c~
E~ P. ~ O P4 ~ .'S

-94- llO-OOlB

h ~i C~
n O
X~
~_CO
:~ ~ U~ O
U~ O ~ ~
:~ o )~ I ~ I I q .4 Q ~
~: a~ ~ I ~ ~ ~ ~ .~J
~0 ~ X

U~
h X
O P. ~::
Q- ~,, o ~ S~ ~ o h Z h ~ ~ ~r H ~ 0 o ~ ~ ~ S O
O 'C Q) ~
U~ ~ U~ O
C~ ~ ~ ~ o ~ Q) c~ ~ ~ ~
z ~ ~ ~ , Q
E~ ~ O ~ a~ I I
Z O ~ X ~ U~
C~
a~
_ ~ O
X ~ ~
¢ ~ ~1 o n o o ,~ ~ Q ~ ~1 Q ~ r~7u~ u) ~ ~ ~3 ~ u~ ~ 5 O ~

t4 O ~ h O

~ O tJ tu~ ~ ~~ ~ tr~
c: o Q) s~ a ~o~
a) I ~ orl ~ ~
~ U~ ~ ::1 t) ~ ~ .~. ~ ~Q~ e qJ ~1 t~ h ~ ~ ~ ~
H hu~ d Q. S ~ ~ O O ~ ~ al ~1 ~ ta o P~ I S
n ~ O Sl ~ ~ ~ O ~ ~dPdP O
:~ ~; ~ E~ O .C ou~o ~n O~ E~
E~ ~ E~ ~ _, -95- ~ llO-OOlB

Table XVII show~s the properties of Group la (alkali metal) polyphosphides of various compositions and physical form. We obsexved that the electro-optical proper-ties are independent of the metal whether it be Li, ~a, K, Rb, Cs, physical form ~ crystal, polycrystal, amorphous (boule or film); and chemical composition, x = 15 or much greater than 15.

~96~ llO-OOlB

~L2~ 2~

P. o ~
C~ >N ~`I
a~ l l Z _~ ~ CO ~ ~ ~ OD
~:. . . . . ~ .
m ~ ,~ r~

E~ r~
l ~ O O
O ~ Or~
E~ a~ ~ , 5~ OOO OOO 0 ~0 Pl V ~I
U~
U~ ~, h .C ~ ~cr~
a) ~ H ~ I I I
o o o o o O ~ ~ ~ _I ~ ~
H Q. ~ a :~CD ,~ ~co r--1~l~ co X _1 ~ o o~oo 000 o o g C~
.,~
~q ~ ~C
~ ~ O Z
E~ O ~ O
E HU~ U~
O E~~3 O O
a~ x ~a~E~s~ ~
O O
x n~ ~ ~ ~cC m ~
U~ Ul ~ U~
3 ~r-l ~1 ,; ,-1 --~ X '1 '~''\ X X
o ~-1 X X ~ ,Xp,X,~
m :z;~ Q Ul h ~ 3; ~ ;z z æP~ v ~d t`q ~ _I
.,1 ., I ~ ~ U~
~,-,, 0 Ul rl r rl rl .

0~ ~: ~ ~0 -~ ~0 ~ :~0 ~ ~10 ~ ~ ~ ~; ~1 0 11~ O 11~ O 11~ O
o ~: ~ ~ ~ O E ~ o ~ ~~ o ~

~97~ llO~OOlB
~,p~ 2:~L
Table XVIII summarizes the properties of mixed polyphosphides and shows those formed of mixed alkali. rnetals have no su~stantial changes in properties î partial. suhsti-tution of ~s on r sites is possible and r,roduces a ~eduction in resistivity and possibly in the band gap (i.e. substi~
tutional doping).

-98- llO-OOlB

~D
~ N . N r-l ~ ~ a~ ~ ~~ . X
04 ~U r-i IH ~1 i h H

H
D
z Ll H
O ~ N N ~ i N N
~1 ~ O C:/ O O O O O

Ui >~
G~ i.~
rl H--i .
~ ~ ~ ~ O~
H h H _ H ~ O
~ ~ ~ ~i X O ~ ~
i i O O o D i i ~~~ ~i r i O o _l t~ ~ ~i C~ O
U~ O U
r-l 31~ ~ ~" ~ii r~)i P~
U~ g Z~ ~ ~ ~ ~ ~ ~ ~ ~ i ~
; z ~1 Q 5~1 N ~I S-l O K KZ Z K ~; K
~: Q~ ai SD
--I X ~ N ~i ,~ N ~i O ~-i p~ , ip, r-i -1 r i C~ I r-i I r'i a~ ~ x ~ s P; ri U~ Q U~ 1~ N u~ h N u~ H
X ~ . z ~ h ~ U ~

-99- llO-OOlB

Table XIX summarizes materials and properties obtained from different starting charye ratios. We find that the best properties are obtained with materials formed Erom starting charge proportions of P to K of about 15 (i.e. between 10 to 30). Below 10 the yield decreases;
above 30 ~he physical properties of the amorphous boules begin to deteriorate.

100- llO-OOlB

TABLE XIX
~P from different startir-g charge~ analyzed x ill Tables I~, X, anh XI above STARIING CHEMICAL X-RAY CONDUCTIV~TY PHOTO-CHARGE ANALY S I S POWDE R ( ohm- cm ) C ONDUCT IV I TY
PATTERN ~ATIO

K/P15 reagent crystalline x = 15 A 10 -10 1o2~1o3 polycrystalline x>> 15 B 10-7-10_9 1o2 amorphous x)> 15 amorphous 10 8-1 9 1o2 R/P15 pure crystalline x = 15 A 10 9 102 polycrystalline x ~> 15 B 10 8 10~
a~orphous x 15 amorphous 10 8 103 crystalline x - 15 A 10 9 102 polycrystalline x~> 15 B 10 9 1o2_1o3 amorphous x)~ 15 crystalline x = 15 A 10 9 10 polycrystalline x ~ lS B 10 8 10 amorphous crystalline polycrystal~ine amorphous x ~> 15 poor physical prop~rties A = pattern similar to crystalline KPi B = p~ttern similar ~o crystalline ~Px -101~ 5;~1L llO-OOlB

We conclude that all these material5 in whatever form have a band gap between 1 and 3 eV, more particularly in a range from 1.4 to 2.2 eV, sinc~ 1.4 eV is the lowest photoconductivity peak we measured and 2.2 eV is the estimated band gap of red phosphorus. The data further indicate that the band gap of the best form of these materials is approximately 1.8 eV. Furthermore, their ~urprising high photocond~ctivity ratio~ of from 100 to 10,000 indicate that they are very good semiconductors.
:
Doping Bul~ amorphous MPX boules obtained by single source vapor transport (Table~ VI, VII, X and XI above) in our three zone furnace having a composition x much greater than can be processed by cutting, lapping, polishing, and etching into high quality, mirror finish wafers of about 0.5 cm diameter.
It is on these samples that we have ~een able to perform electrical measurements with diferent geometrical -~
arrangements of electrical contact~ to determine acç~rately the bulk conductivity of the materials~ ~y 2 probe and 4 ~-probe measurements, we ascertained the bulk conductivity of these materials to be 10 8 to 10 9 (ohm-cm) 1~ Thîs conductivity is too low for the material to be able to form .-a sharp junction with r~ctifying properties. Therefore, it was our aim to find a foreign element tdoPant) which would -.
affect the conduction mechanism in ~he material and increas~ -conductivity. As is typical of other amorphous semiconductors, the presence of small amounts of impuritie~ ~
in th~ material do not affect the conductivity and~ above :.
room temperature, we find intrinsic behaviox with an activation energy e~ual to approximately half he bandgap9 indicative of a midgap Fermi level~ The low conductivity a~d large photoconductiVitY ratio indicate a small number of dangling bonds. This indicates that a strong perturbation -102- llO-OOlB

of the elcctronic wave function of the P-P bond will be required to modify the conductivity and conductivity type.
Two approaches were taken: ~1) substitute As or Bi into the P site; (2) diffuse a foreign element into the amorphous matrix, In the first method K/As2/P]3 has As incorporated into the matrix. The conductivity ie; increased by 2 orders of magnitude (Table XVIII3, and the material remains n type.
In the second method, after txyin~ many conventional diffuser~ (e.g. Cu, Zn, Al, Tn, Ga, KI) in vapor, liquid and solid phase diffusion with no success, we found a surprising success with the diffusion of Ni and then Fe and Cr from the solid phase. For example, a layer of Ni was deposited by vacuum evaporation onto a well prepaxed surace o a high x, KPX wafer. After annealing for several hours~ the Ni was found to diffuse for about 0.5 micrometers into the sub- :
strate and the conductivity increased by more than 5 orders o magnitude. The conductivity is still n type.
More speciically, 1500 angstroms of Ni were deposited onto the wafer in a Varian resistanc~ heated vacuum evaporator under pressure of 10 ~orr. The sample wa~
sealed in an evacuated Pyrex tube and heated for 4 hours at 350~C. The top Ni layer was removedD The conduotivity--~-~
measured by the two probe method showed an increase from 10 ~ to greater than 10 4. Eleotro spectroscopy or chemical analysis (ESCA) depth profiling of the sample showed the diffusion depth to be 0.4 micrometers and the -chemical bonding of the Ni to be N~9 i.e. ree Ni in the material. The wavefunction of the Ni overlaps with electronic wavefunctions in the P-P matrix, affecting the ~ ~
conduction ~mobility). The Ni concentration is ~reater than ~, about 1 atom percent.
Evaporated gold top contacts or dry silver paint in ~
coplanar f~shion form ohmir contacts to the doped layer.
VariationS in the diffu~ion temperature show 350C to be optimum for Ni diffusion, Variation in the diffusion ~ime follow the diffusion equation ~diffusion depth is proportional to square root of ; 3 :

., -103~ 110-001B

time) and 1500 angstromS of Ni heated at 350C or 60 hours, showed diffusion depth of 1.5 micrometers as measured by ESCA. 350C approaches the highest temperature these amorphous materials may be subjected to~ -Ni diffusion can also be accomplished from the liquîd phase, ~uch as from a Ni-Ga melt, ox from the vapor phasa, such as from Ni carbonyl gas.
It was further found that Fe and Cr show similar ~;
behavior under the above processing procedures.
For example, we ~ook a cut wafer from a bulk amorphous high x boule obtained by the single source vapor transport and evaporated 500 angstroms of iron o~to it and then diffused it into the wafer at 350C for sixteen hoursO
Applying two pressure probes to the doped material gave a full non-linear characteristic on the Tektronix curve tracer.
On another wafer of high x material we evaporated 300 ~ -angstroms of niokel and 200 A of iron, then heated the wafer to 350C for sixteen hours. We then evaporated two 1 mm radius aluminum contacts 2noo angstroms thick and measuring the current voltage characteristic with the Tektxonix curve tracer between the aluminum dots, ayain obtained a full non-linear characteristic.
On another wafer of high x material produced by single source vapor transport, we e~aporated 500 angstroms o~
nichrome and then heated the wafer or dif~uslon at 350~C
for sixteen hours. We then evaporated two aluminum 1 nun -radius dots 2000 angstroms thick onto the wafer and again measured ~ full non-linear characteristic between the two aluminum dots.
We thus conclud~ that nickel, iron and chromium are ;
useful diffusants in these materials for lowering conductivity and that on the lower conductivity material junctions can be effected with wet 5ilver paint, pressure contacts and aluminum contacts.
Other elements besides Ni, Fe and Cr with occupied d or f outer clectronic levels that can overlap with the phosphorus levels are expected to be able to affcc~ the *Trademark ' -104- ~f~ 110 OOlB

conductivity in these materials such as to give p-type material and foxm p/n junctions for solicl state devices.

Amorphous High Phosphorus Material By Two Source Vapor Transport Two types of materials were obtained by`this method and the properties of these were investigated.
1) Amorphous bulk KPX, (Example VI) where x equals approximately 50 on one side and x is much grèater than 15 on the other. Surface analysis supports the hypothesis of the template effect, which is very strong in this instanc The surface of a cut and polished sample is of very high quality, low number of defects and voids, uniform etching pattern.
The conductivity measured was by the two probe technique 10 15Ohm-cm) 1 and the photoconducti~ity ratio under illumination of 100 mW/cm is greater than 103. The photoconductivity peak is approximately at 1.8eV, indicating a bandgap of that order. The data indicates that the P-P
bond dominates th~ electrical and optical properties o~ this material as well as those in Tables XVI, XVII, XVIII and XIV, and its ~trong photoconductivity ratio is consistent with a highly r~duced level of dangling bonds.
~ ) Amorphous thin films of KP15, (Reference ~o. 47 Table XII) deposited onto glass slides which have a metal layer deposited on them for a back contact to the thin film~
The success in the thin film deposition of RP15 opens the opportunity to manufacture many types of thin film devices.
The amorphous RP15 thin films deposited by the 2 source technique have a thickness of approximately 0.5 micrometers over an area of 3 cm . The film is uniform and the surface roughness does not exceed 2,000 angstroms.
The film is chemically stable. Figure 25 is a photomicro-graph at 2000 magnification of the surface of one of these KP15 films. ~he adhesion to the substrate is excellent.
Quantitative analysis of the film was performed using a Scanning Electron Microscope (SEM) and an Energy Dispersive -105- llO-OOlB

X-ray ~EDAX) measurement. The composition of the film was found to he in agreement with the KP15 nominal composition.
The uniform compo~ition, homogeneity, and pinhole free sur-~ace leads to uniform electro-optical properties across the films.
In view of the diffusing capability of Ni into bulX
amorphous KPX, an Ni film 172 was evaporated onto th~ glass substrate 170 to form a back contact for the amorphous KP15 layer 174 as shown in Figure 26.
The Ni serves as a back contact and a diffuser. ESCA
and SEM profiling shows Ni to dif~Euse significantly into the KP15 film 174 at a ra~e of 200 angstroms per hour during the KP15 growth process.
In more detail,we deposited by vacuum evaporation lSOO
angstroms of Ni 172 onto a glass slide 170 at 10 ~orr.
pressure. Part of the Ni surface is then masked with a Ta mask in order to have a material free zone for electrical contact. -Two micrometers of amorphous KP15 174 is deposited in our two source apparatus onto the Ni film 172. The composi-tion of this film has been identified to be KPl~ r it ;S
amorphous and has more than 1~ Ni diffused into the film.
Pressure contact with an electrical pro~e was applied - -~-~to the top of the KP15 film. The two leads 9 rom the baok contact and the top pressure contact, were connected to a Tektronix Curve Tracer 176 to observe the current voltage characteristics. The forward characteristic of the recti~
fying pressure contact junction is shown in Figure 27, which indicates a junction with a barrier height of O.SeV and current in the mA range.
As shown in Figure 28, we also deposited by vacuum evaporation a 2 mm radius Cu rontact 178 onto the top surface 180 of ,a KP15 amorphous layer 182 grown ~y the two source techniqule on a Ni layer 184 deposited on a glass substrate 186. We connected the Tektronix curve ~racer 176, as shown, and measured ~he full orward and reverse biased .:

~p~
-106- llO-Ools junction curve shown in Figure 29, which thus mdic~ed that cu forms junctions with these materials.
Subsequently, smaller metal dots were deposited as top contacts in order to reduce the effect of leakage currents at the edges of the contacts. 10 3cm area top contacts and 10 5cm2 top contacts were deposited in the vacuum evaporator through mechanical masks. The I-V
characteristics shown in Figure 31 were obserVed with Cu, Au, and Al top contacts. They appear as the breakdown voltages of two back to back diodes in each instance.
Similar curves were obtained with Ni, Ti, Mg, and Ag as the top contacts.
The most significant difference appears in the fact ~-that Au contacts change the I-V characteristic after applying lOV to the device. The I-V characteristic become asymmetric, as shown in Figure 32, and a more ohmic contact is formed at tha Au interface after this "forming" process.
The "forming" is consistently observed with Au, and intermittently observed with Rg and Cu top contacts. The "forming" does not permanently affect the device, but it reappears every time a voltage is applied. ~eating the device at 300~C does not affect the phenomenon~ Cooling the device to -20~C results in very sharp I-V characteristics ~Figure 33~
It appears as if the "forming" may be a breakdown of a high resistance layer remaining between the diffused part of the device and the top contact~ Capacitance - voltag~ (C-V3 -;-characteristics shown in Figures 34, 35, and 36 point in the ~;
same direction. Al and Au top contacts have C-V
characteristics of double diode5~ but conver into single diode behavior in the case of Au contacts. If we assu~e a dielectric cons~ant of approximately 10, we can extract a carrier ~oncentration of approximately 1016 c~rriers per cm3 near the ~unotion and a carrier mobility of approximately lo~2 to lcm~/volt ~econd. Frequency dependence of the capacitance ~nd resistance in Figure 37 can be used to model ~he multiple ju~ctions that can form in such a structure wi~h a graded diffusion profile i~ the -107~ 5 ~ ~ 110 001~

active material. In addition, poor bu:Lk material quality (low density) and rough surface morphology could contribute `~
to the complex observations. Nonetheless, junction formation capability on amorphous ~o source thin film KP15 has been demonstrated.
Some of the above phenomena, such as :'forming" wilth Au top ~ontacts was also observed with flash evaporated thin films deposited on Ni. This film is not pure KP15, but has excellent quality. No C-V clependenc~ was seen in this case. The device, which was very thin, had a good rasponse to light and a small ~10 6amps) current was drawn from it under short circuit conditions when illuminated with visible light.
We expect that KP15 thin films made by CVD technique will result- in similar behavior when the films are sufficiently thick. At the moment they have been too thin and have been found to short out.
The formation of junctions with these materials indicates that they may be u~ilized to form pn junctions, Schottky diodes, or Metal Oxide Semiconductor (~OS) devices. - ~
We expect that by utilizing the above noted class~s of ~ ~
dopants, that the materials can be converted to p-type conductivity and thus will be useful in the entire range of -semiconductors. -~
The photoconductivity ratio was obtained in all these by forming a semiconductor device comprising our material and means attached to the material for electrically communicating with ito This means comprised two single ~-electrodes 80 and 82-attached to the material as illustrated in Figure 30.
More specifically, for a single crystal of MP15, two copper strips 80 and 82 were adhesively attached to a glass substrate 84. A samplé86 of KP15, made according to the above teachings, was bridged across strips 80 and 8~ at one end thereof and attached thereto by sil~er paint 88.
Meter Q0 attached to the opposite ends of strips 80, 82 introduces an electrical potential to the RP15, and thereby permi~s men5uremènt of the resistivity of the ~P15.

" " . v~ 1 -108~ 2 ~ llO-OOlB

The rcsultant device of Figure 30 and similar devices using our other materials established that our high phosphorus materials can in fact be used to control the flow of electrical current, at least as a photosensitive resistor.
In addition, our materials show luminescence character-istics with an emission peak at 1.8eV at temperatures of four degrees K, and luminescence at ambient temperatures.

Preparation of Large Crystal Monoclinic P

Rubidium We have found that the RbP15 can be utilized to produce large crystal monoclinic phosphorus.
A 0.62g sample of RbP15 encapsulated, ln vacuo, in a lOmm O.D. x 6mm I.D. x 5.0cm quartz tube was vertically positioned in a crucible Eurnace and subjected to a tempera-ture gradient such that the ~bP15 charge was maintained at 552C while the top of the tube was maintained at 539~C.
After heating for approximately 22 hours, the tube was open-ed and single crystals of monoclinic phosphorus, as larg~ as 3.0mm on edge, in the form of truncated pyramids were found in the upper (cooler) region of the tube.
We found that large crystal monoclinic phosphorus can also be prepared from mixtures of Rb and P in the atom ratio of 1 ~o 15 5RbP15).

Cesium and Sodium Large single crystals of monoclinic phosphorus were also grown via vapor transport using either CsPlS or NaP15 charges formed in our condensed phase process. In each run approximately 0.5g of the appropriate alkali metal polyphosphide was sealed in vacuo in a quartz tube (lOmm O.D. x 6mm I.D.) of length 8.9cm. The tubes were then subjected to a temperature gradient su~h that the alkali metal polyphosphide charges were maintained at 558C while the tops of the tubes were maintained at 514C. After 48 hours, large deep-red crystalline stacked square platelets of monoclinic phosphorus formed from the CsP15 charges.

-ln9~ llO-OOlB

The morphologies of the monoclinic phosphorus crystals grown from CsP15 and NaP15 condensed phase charges appear to be v~ry similar, that is, stacked square platelets. ~his i5 in contrast to the truncated pyramidal habit of the monoclinic phosphorus crystals grown from a RbP15 char~e.
We found that large crysta:L monoclinic phosph~rus can also be prepare from Cs/Pll, and Cs/Pll and Cs/P15 mixtures maintained at high temperatures.
Potassium Using similar processes we have also produced mono-clinic phosphorus crystals from condensed phase KP15, and from mixtures ~f K/P30 and K/P125.
Lithium No experiments have been conducted with lithium/phosphorus charges. However, we expect that large crystal monoclinic phosphorus can be prepared from the materials under similar conditions~
Effect of Temperature While the nature of the alkali metal present seems not to be important, the temperature at which the charge is maintained is apparently very important to the crystal growth process. In the ca6e of the Cs~Pll ball milled system, large crystals were produced in experiments where the charge was maintained at 555C and 554C. However, in experiments where the charge was held at 565C and 545C, no large monoclinic crystals were produ~ed.
Referring to Figure 38, using our preferred apparatus, we sealed a 0.6gm sample of RbP15 prepared by our condensed phase process in vacuo in a 12mm OoD~ x 6mm I.D. x 8cm long glass tube 270. The top was ~ealed with a 16mm diameter flat glass surface 272. Fill tube 274 is provided with a constriction 276 at which it is sealed after charging and evacuation.
The tube was subjected to a temperature gradient such that the flat surface 272 at the top of the tube was main-tained at 462~C, while the charge at the bottom of the tube -110~ S 2 ~ llO-OOlB

was maintained at 550C. Af~er heating for 140 hours ap-proximately half of the original charge had been transported to the flat surface.
The resulting button-like ~oule was cleaved and exam-ined. It was made up entirely of uniform light-red fibers -- not the desired large crystal monoclinic phosphorus, Figures 44 and 45 are SEM photomicrographs of this product at 200 and lOOOx magnification,respectively~ :
The SEM photomicrographs of Figures 44 and 45 proved to be a surprise. The individual ~fibersW consist of bundles of long platelets which are attached such that they appear to be star-shaped rods when viewed from the end. This material is thus quite different in appearance from the "twisted tube" fibrous phosphorus produced via vapor transport from a 99.9999~ red phosphorus charge (see below).
We conclude that the condensing temperature to form large crystal monoclinic phosphorus should be in the range ~:
of 500 to 560C. Further experiments indicate that the :~
preferred condensing temperature is about 539C. -- :
The charge must be heated to a temperature above 545C :~:
and below 565C as previously indicated. Our preferred range is 550 to 560C with about S55C giving the besk resuI~s.
. , .-'~
Effect of Composition We have produced monoclinic phosphorus from charge :- `
ratios of P to alkali metal of 11 to 125. ~owever,a ratio - ~:
of about 15 seems to work best. --;-Characteristics vf Monoclinic Phosphorus Co~densed from Vapor in thë Presence of an_Alkali Metal : :
Figure 39 is a photomicrograph at SOX magnification showing a pyramidally shaped monoclinic crystal of phos- ~
phorus prepared from a RbPl~ charge. These crystals are .:
hard to cleave. Similar crystals are produced fro~ charges utilizing sodium as the alkali metal. We have produced crys~als as large as 4 x 3 x 2mm. ~ .

.. .. . . .

~ llO-OOlB

Figure 40 is a photomicrograph, at 80X magnification, of a crystal of monoclinic phosphorus produced from a ball milled mixture of Cs/Pll. These platelets are easy to cleave int~ mica-like sheets. Similar crystals can be pro-duced from a charge of K/P15O We have produced crystals in this habit as large as 4mm on a side and 2mm thick.
We have determined tha~ the crystals are birefringent.
When placed between crossed polarizers in a polarizing mi-croscope, they rotate the light and allow some of it to pass through. Thus they may be utilized as birefrinyent devices such as optical rotators in the red and infxa-red portion of the spectrum.
Chemical analysis indicates,that they contain anywhere from 500 to 2000 parts pex million of an alkali metal. They are made in a process which takes as little as ~2 hours versus the 11 days employed in the process of the prior art ~ -~
to produçe Hittorf's phosphorus. ~, The powdPred X-ray diffraction pattern of these crys~
tals is consistent with that of the prior art Hittorf'~
phosphorus. , The photoluminescence spectra shawn in Figures 41 and ~' , 42 were taken with an Argon laser Raman spectrometer. A ~' broad peak at 1.91 eV is clearly observed with a half width '"''h o about 0.29 eV. This indicates a band gap of about 2~0 eV '~
at room temperature~
The Figure 41 spectrum was taken utilizing a monoclinic ~
crystal of phosphorus prepared in the presence o cesium, ' , while the Figure 42 spec~rum was ta~en using monoclinic phosphorus condensed in the presence of rubidium. , ~, The Raman spectrum of Figure 43 was taken utilizing a monoclinic phosphorus crystal formed i~ the presence of Rubidium. The peaks 280, 282, 283, 284, and 285 are a~ wave ,' numbers 285, 367, 46S, 483, and 529~
Evaporated dots about 25 micrometer5 in diameter werR
deposited on large crystals of monoclinic phosphorus (rom a Rb/Pl~ source) for electrical measurements~ The re~istance of the crystals was found to be 10 ohm to J
107 ohm and practically in~ependent of the geometry of the f ' .. . .

-112- llO-OOlB

crystal and the si~e of the contacts. This reflects surface resistance.
These crystals may be utilized as the substrate for depositing 3-5 materials such as Indium Phosphide or Gallium Phosphide. They may be utilized as phosphors in luminescent displays, semiconductors, lasers, and as starting materials for other semiconducting devices.

Twisted Fiber Phosphorus The presence of the alkali metal in he charge appears to be critical to the production of large crystal monoclinic phosphorus. We attempted to produce large single crystals of monoclinic phosphorus from 99.9999~ pure red phosphorus by mimicking the conditions used successfully with the -~
various alkali metal/phosphorus systems. This attempt failed. No monoclinic phosphorus was produced. For exam ~ ~
ple, a 0~6g sample of 99.9999% pure red phosphorus was heat- ;-ed at 552C in a sealed evacuated tube in a vertically positioned 10mm outside diameter x ~mm inside diameter quartz tube. The temperature gradient between the bottom ;
and top of the two and three-quarter inches long tube was 43C. After heating for 24 hours, more than half o the charge had been transported to the top third of the tube where a boule had formed.
Surprisingly, the boule was found to consist entirely of a red fibrous material. Several long (approximately 1.5mm) fibers ~ere found in the vapor space at th~ bottom of ~ `
the boule. Microscopic examinatio~ of the deep red fibers revealed that they are twisted. ~;
XRD data secured on the fibrous material were found to match those secured earlier on polycrystalline KP~ where x is much greater than 15. Figure 46 is an SE~ photomicro~
graph at 500 magnification of these fibers.
Differential thermal ana~ytioal data was ~ound to be similar to that secured on polycrystalline high x material.
For two DTA determinations, the first heat plot consists of a single endotherm at 622C (average). ~he second heat plot -;
consists of a single endotherm in both cases - at 599C.

S~L
-113- llO-OOlB

The DTA data secured earlier on polycrystalline high x ma-terial ~onsists of a first heat single endotherm -- at 614C
and a second heat single endotherm -- at 590C. Thus, we observed substantial similarities between the fibrous phos-phorus pre~ared from 99.9999~ red phosphorus and polycrys-talline high x material.
Flash Evaporation We have succeeded in formincl stable thin film amorphous coatings on glass and nickel coat:ed glass substrates using a flash evaporation process. The flash evaporation apparatus is generally indicated at 302 in Figure 47. It comprises a glass cylinder 304 connected to a vacuum system ~not shown) through tubing 306. Argon is supplied at inlet 308 of sup-ply tube 310. Reservoir 31~ is filled with powdered KP15 formed by the Condensed Phase method. It is agitated by means of a vibrator ~enerally indicated at 314 and picked up ~-by the flow o Argon gas through the venturi generally indi-cated at 316. It then flows into the reactor 304, passing through tube 317 into a steel susceptor 318. The susceptor ~
is heated by means of a RF coil 319 to a temperature of at ; -least 900C,which causes the RP15 to vaporize. At the end of tube 317, as shown in Figure 49 , a nozzle is formed by incorporating a plurality of small tubes generally indicated at 320 in Figure 48, having a plurality of small oriices -~
321. Tubes 317 and 320 are alumina and tubes 320 are held -`~
within the end of tu~e 317 by means of magnesium oxide ce-ment 322.
The XP15 upon v~porization dissociates int~ its -constituents and thç vapor ~s carried ~y the Argon gas through the orifice~ 321. The film is deposited on a cooler substrate 324. The substrate may be heated by means of hot wires 326 fed by electrical connections 328.
Alumina tube 317 has a one-quarter inch outside diame- -ter and one-eighth inch inside diameter~ Tubes 320 have a `~
one-sixteenth inch outside diameter, are one-quarter inch long and have four one-sixteenth inch diameter holes ~ ?
throughthem.
:

~X~5 ~
114- llO-OOlB

The apparatus is operated under a vacuum of 0.1 to 0.5mm Hg. Amorphous films of up to 1 micron thick may be formed in runs of up to fiEteen minutes. At the end of a run, the substrate 324 reaches a temperature of 200-300C, depending on whether it starts out at room temperature or is initially preheated to 200C~
Chemical Va~or Deposition We have prepared thin ~ilms of KP15 by means of Chemical Vapor Deposition.
A typical chemical vapor deposition reactor is shown in Figure 50. It is constructed of Pyrex. The reactor chamber 401 is a 26 mm I.D. x 27.0 cm long tube in the center of which is positioned a 6.0 mm I.D. x 30.0 cm long tube 402 which serves as both a thermowell and substrate holder. The thermowell is held in position by an adjustable O-ring collar 403. The vent tube 404 allows for the continuous removal of the gaseous exhaust stream. It is attached to a trap (not shown) which removes the unreacted phosphorus before venting of the stream to air. The vent tube 404 and O-ring collar 403 are attached to the reactor chamber 401 through a 2.0 cm I.D. O ring joint 405. The reactor cham~er 401 is located in a resistance furnace generally indicated at 406.
Molten phosphorus is metered by a piston pump ~not shown) through a 1.0 mm I.D. capillary tube 407 into a vaporization chamber 408. The molten phosphorus is evaporated in the vaporization chamber 408 by a stream of argon which is injected into the vaporization chamber 408 through the 6.0 mm I,D. inlet tube 409. The gaseous phosphorus/argon stream enters the reactor chamber through nozzle 410. The no~zle 410 has an opening of 4.Omm. The evaporation chamber 408 is located in a resistance oven generally indicated at 411.
A gaseous mixture of potassium and argon is metered into the reactor chamber 401 throu~h inlet tube 412 which has a 6.0 mm I.D. ~eat argon, which acts as a shroud for the potassium/argOn stream, Pnters the system through $.O mm I.~. tube 413. Both the potass~um/argon stream and neat --115- ~ Z~ llO-OOlB

argon stream enter ~he re~ction chamber 401 at 414. The potassium/argon and neat ar~on lines (412, 413~ are located in a resistance oven generally indicated at 415.
The substrate 416 is positioned on the thermowell ~02.
The temperature of the substra~e 416 is determined by A
thermocouple 417 posi.tioned direc~ly below the substrate 416 on the thermowell 402.
During operation, ovens 406, 411, and 415 are maintained at appropriate temperatures. The gaseous reactant st.reams enter the reactor chamber at 410 and 414.
The exhaust gas mixture leaves t:he reaction chamber through the vent tube 404. The desired film forms on the substrate 416.
The substrates are maintained at a temperature of 310-350C, the temperature being maintained oonstant to plus or minus 2C.
In a typical run 1.24g of white phosphorus and 0.13g of potassium are delivered into the reactor over a two hour period. The total Argon flow rate is maintained at 250ml per minute during the run.
A number of experiments were conducted in which phosphorus~argnn and potassium/argon were fed simultaneously into the reactor. The phosphorus/argon stream was maintained at approximately 290C and the potassium/argon stream at approximately 410~C. The calculated atom ratio of reactants in the reactor was P/K approximately 15. The reactor was maintained at 300-310C7 In a typical experiment the liquid phosphorus feed rate was 0,34ml per hour.
Amorphous KP15 films were preparea using nickel-on-glass substrates. The films are about 0.3 millimeters thick.
With a run time of 1.0 hour, the films produced were found to have nominal KP15 composition. The thickness of the film was dependent on the position. of the particular substrate in the reac~or. Examination of the films usin~
SEM showed them to be quite uniformO

--116~ llO-OOlB

Purification of Phosphorus 50 grams of Atomeryic phosphorus, "99.95% pure", was subjected to a 450-300~C gradient for 75 days. After this admittedly very long time, 21% of the material remained behind and 60% of the charge ended up as amorphous, bulk deposits.
Previous analysis showed the! Atomergic phosphorus to be less than 99.90% pure, probabLy closer to 99 . 80% t with aluminum, calcium, iron, magnesium, sodium, and silicon as major impurities (all greater than 0.01%, and some greater than 0.05%). This material costs about $2~0 per kilogram.
In comparison, "99%" P from Alpha Ventron is $17/lb., or $37.5/Kilo.
Table XX summarizes the results of flame ~mission spectroscopy on three materials generated ~y the aforementioned treatment.

-117 ~ 2 3L l l O - O O l B

.,~

,1~ .
~P
~ ,C
.,, o ~ ~ o s~
s~ o o I
~ a: ~ O O ~q O O u~ ~ o J~ Ir~ r ~In I o * ~ o I I ~ I Ia) o I
U~ ~, N N t~ 1 ~ O
S~
:~
m ~C U~
X ~ d~
~ W rl O ~ O O d~ 0 0 m ~ ~ , O o ~
~ ~ ~, N ~ O N ~ O O O ~> O I O
E-~ :~ ~ o I ~D IO ~ ~ ') I ~ ~71~ ~I ~ a E~ ~ o ~ o ~ D J C~ V I h ~ ~O el~ ~D ~ O ~~D ~ ~ ~ O ~, ~1 O

P. Q~ ~
~: O ~ Q) O P~ h k X
~ ~ . ~ . O Ei h ~ O O O ~ c~ Q) `
Q) I O O I ~ I C:4 1 ~ G~ ~ C~ D ~ O I ~r o o g~
O C~ O tD I I ~ I O ~D ~ O ~D ~ Q, ,~_ I o ~r o ~
o ~ ~ . er ~ O ~ O O ~D ~ O ~ p~ O
u~ O E~
.q o o ,1 ~ O O

a) .,", .
~ a~ æ z z ~

-118- ~J~ 110-OOlB

Material ~ was a residue, dark brown in color, throughout the charge zone, which did not undergo vapor transport~ ~terial designated Material B was a hard boule of material, li~ht in color, which did not vaporize, primarily because its position in the charge zone results in it being at a slightly lower temperature than 450C~
Material C was an amorphous boule in the cold zone.
Clearly, most of the impurities of the charge remain in Material A at fairly concentrated amounts. The high impurity levels would be expected to give rise to a lower vapor pressure of phosphorus at a given temperature. The impurity level of the materia:L pretty well reflects the values for the initial charge material. The boule, Material C, is a pretty pure material, with the 50dium content being the major observed contaminant. Taking the sum of contaminants, at their maximum indicated levels, this material has a purity level of 99.9g7~, at worst. The comparable material, obtainable from commercial sources, as 99.999% P, costs abou~ $1,800/kilogram.
Clearly, we have illustrated a cost effective method for purifying red phosphorus to a high degree.
Reinforced Materials The use of phosphorus compounds as fire retardant additives is well known. Because of the highly stable nature of the alkali metal containing phosphorus materials disclosed herein, they may be utilized for such purposes.
The fibrous and plate-like forms of the materials disclosed herein, for example, fibrous KP15 and KPX
where x is much greater than 15p the plate-like monoclinic phosphorus habit, the twisted tube form of phosphorus~ and the star-shaped material of Figures 44 and 45, all show promise as reinforcing additives for plastics and glass.
The twisted tube fibers and star-shaped fibers should be of particular value due to their ability to mechanically interloc~ with the matrix of a composite material. Their fire retardant qualities should also prove useful in such materials.

-~19~ L5~ o. ~ol~

coatillgs As previously discussed, many of the materials disclosed herein form high stable amorphous coatings, with good adhesion to metal and glass. The MP15 amorphous films are particularly stable and provide good adhesion to metal and glass. Thus they may be utilized as corrosion inhibiting coatings on metals and as optical coatings on glass.
A coating of approximately 1000 angstroms on an appropriate infrared optical component such as germanium, will provide a window transparent to infrared but showing absorption in the visible. Such coatings when combined with coatings of other materials of differing optical indexO can be utilized to provide anti-reflection coatinys on infrared optics.
Our experiments show that MPX material can be deposited as films with good adhesion on steel, aluminum, and molybdenum. The films are ductile, non-porous, polymeric, and non-brittle.
Thus the materials shown herein should find wide application as coatings and thin films.

-1~0- llD-OOlB

I~JDUSTRI~L APPLICATION

Thus we have disclosed an entirely new class of high phosphorus semiconductor materials. These semiconductors comprise catenated covalent atoms where the catenated covalent bonds serve as the primaxy conduction paths in the materials. The catenated atoms form parallel columns as the predominant local order. Preferably the atoms are trivalent and have bonding angles that permit tubular, spiral or channel-like columns. Th~ columns may be joined by atoms of one or more different elements bonded to two or more of the catenated columns.
We have disc:Losed in particular high phosphorus and mixed pnictide semiconductor materials of this class. These include high phosphorus polyphosphides of the formula MPX
where x ranges from 7 to 15 and entirely new materials where x is very much greater than 15 -- for all practical purposes pure phosphorus~
These mat~rials can be characterized as containing groups of seven or more atoms organized into pentayonal tubes. They may be characterized as having the formula MPX where x is greater than 6 and they may be characterized as comprised of phosphorus in a molar ratio o~
phosphorus to any other atomic constituent greater than 6;
they may be characterized as high phosphorus materials where the phosphorus atoms thereof in substantially all local orders comprise phosphorus atoms joined together by multiple covalent p-p bonds organized into layers of all parallel pentagonal tubesO They may be charac,erized as polyphosphides containing alkali metal atoms wherein the number of consecutive covalent phosphorus-to-phosphorus bonds is sufficiently greater than the number of non-phosphorus-to-phoSphorUs bonds to render the material semiconducting. They may be characterized as having a skeleton of at least seven covalently bonded phosphorus atoms having associated therewith at least one alXali metal atom, conductively bridgin~ the phosphorus skeleton of one -121- llO-OOlB

unit with the phosphoxus skeleton of another unit; they may be charactcrized as a polyphosphide having the formula MPX
where M is an alkali metal and x i5 at least 7.
These materials may be furt:her characterized by having a band gap greater than 1 eV thatis, from 1.4 to 2.2 eV, and for the best materials we have discovered,approximately 1.8 eV
They may be characteri~ed by having a photoconductivity ratio greater than 5; more particularly, within the range of 100 and 10,000.
These materials may be further characterized by their trivalent dominant atomic species; the homatomic bonds formed by the dominant species; the covalent nature of these bonds; the mate~ials~ coordination number of slightly less than 3; the materials' polymer nature; their formation in the presence of an alkali metal or metals mimicking the ~ . :
bonding of alkali metals with the dominant species, in the crystalline form, their pentagonal parallel tubes either all : :~
parallel in KP15-like materials, paired parallel crossed . ~.
layers in monoclinic phosphorus, cr all parallel twisted tubes in twisted fiber phosphorus; their ability to form amorphous films and boules retaining their electronic qualities; and by their methods of manufacture; and other qualities made apparent in the preceding description. ~-The amorphous materials W2 have di5covered maintain the electronic qualities of the KP15 all parallel pentagonal tube structure and in theory, at least, it appears that that -~
structure is maintained in the local scale in our amorphous materia~s. However, we do not wish to be bound by any :-particular theory in this matter. In particular, the claims appearing below should be interpreted broadly to cover all ~:~
aspects of our invention, regardless of later acquired knowledge that miqht be said to conflict with the theories and hypothesis we have put forth, both explicitly and implicitly . . '~
We have disclosed junction devic~s, photoconductive -~resistive) devices~ photovoltaic devices and phosphors made from these materials.

.- .

-1~2- 1 21 ~t~2 1 llO-OOlB

We have disclosed resistance lowering dopants, namely Nickel, Chromium and Iron, leading to the conclusion that substantially the entire group of atomic species having occupied d or f outer electronic levels may be utilized if of the appropriate atomic si~e.
~ e have disclosed resistance lowering substitutional dopins with Arsenic which indicates that all Group 5a metals may be utilized.
~ e have disclosed junction devices having a back contact of Ni, Ni diffused therefrom, and top contacts of Cu, Al, Mg, Ni, Au, Ag, and Ti.
We have disclosed new forms of phosphorus wherein the local orders are all substankially parallel pentagonal tubes, twisted fiber phosphorus and monoclinic phosphorus.
We have formed these new forms of phosphorus by vapor deposition. The all parallel and monoclinic form require the presence of an alkali metal during deposition.
We have disclosed both amorphous and polycrystalline films of MP15 where M is an alkali metal. We have con-structed various semiconductor devices from all of the all parallel pentagonal tube mat~rials, including wafers of MPX where x is much greater than 15, including a new form of phosphorus, amorphous thin films of ~P15 and æmorphous thin films of KPX.
We have disclosed me~hods of maki~g metal polyphos-phides and two ~ew forms of phosphorus by controlled two temperature single source techniquesO
We have disclosed methods o making our high phosphorus materials by two source vapor transport.
~ e have disclosed a method of makin~ high purity phosphorus.
We have disclosed methods of making crystalline and amorphous forms of MPX where x ranges from 7 to 15 by con-densed phase methods~
We have disclo5ed chemical vapor deposition, flash evaporation and molecular flow deposition methods~

5~
-123- 110-001~

Industrial applicatio~s of the semiconductor materials and devices we have discovered are manifest, running the whole gamut of semiconductor applications. The crystalline materîals may be used as reinforcing fibers and flakes for plastics, glasses and other materials. The materials of our invention may be used a~ coatings o~ me~als, glass t and other materials. The coatings may protect a substrate from fire, oxidation, or chemical ,a~-tack, The coatings may be employed or their infrared transmitti~gO visable liyht absorbing qualities. They may be employed with other materials as antireflection coatings on infrared optics.
The materials may be used as fire retarda~t fillers and coatings. Monoclinic phosphorus may be used an an optical rotator.
It will thus be seen that th~ objects set forth above iamong those made apparent rom the preceding description ara efficiently attained and that c~rtain changes may be made in carrying out the above methods, processes and in the ab~ve articles, appara$us and products wi~hout departing from the scope of the invention. It is intended that all matter contained in the above description shall be interpreted as illustrative and not in a limiting sense.
It should be understood that w~ have used crystalllne to mean single crystals or polycrystalli~e material unless otherwise stated~ ~morphous as dis inct from single crystal ~-~
or polycrystalline, means i~morphous to X-ray diffraction.
All periodic table references are to the table prin~ed on the inside front cover of the ~Oth edition of the ~andbook of Chemistry and Physics published by the CRC Press Inc., Boca Raton, Florida. Alkali metals are identified therPon and herein in Group la and pnictides in Group 5a. All ranges stated herein are inclusive of their limits.
By s~miconductor de~ice we mean any device or apparatus utiliæing a semiconductor material. In particuliar~
semiconductor device i~cludes Xerographic surfaces and phosphors regiardless of how they are excited, as well as photoconductors~ photo~oltaics, junctions, transistors, integrated circuits and the like.

-124~ 3 ~2~ 11O-OO~B

It is also to be understood that the following claims are intended to cover all of the generic and specific features o~ the invention and discovery herein described and all statements of the scope thereof which as a matter of language might be said to fall therebetween.
Particularly it is to be understood that in said claims, ingredients or compounds recited in the singular are intended to include compatible mixtures of such ingredients whenever the sense permits.
Having described our inventions and discoveries what we claim as new and desire to secure by Letters Patent is:

Claims (252)

CLAIMS:

1. A solid film material of MPx deposited on a substrate where M is one or more metals, P is one or more pnictides and X is at least 7.

2. A new form of solid phosphorus material wherein the phosphorus atoms are joined together by multiple covalent phosphorus-to-phosphorus bonds forming substantially pentagonal tubular arrays of bonded phosphorus atoms the local order substantially throughout said phosphorus being defined by said tubular arrays having predominantly all of their axes generally parallel to each other.

3. A solid stable material having the formula MPx where M is one or more alkali metals and P is primarily phosphorus but may include one or more other pnictides and x is greater than 15.

4. A high phosphorus material formed as the deposi-tion product of vapor transport from separated heated sources of phosphorus vapor and the vapor of one or more metals.

5. High phosphorus, polyphosphide film materials formed by chemical vapor deposition.

6. High phosphorus, polyphosphide film materials formed by flash evaporation.

7. A semiconductor device formed of material substan-tially comprising throughout, atoms characterized by the vast majority of the atoms of said species having three, homatomic, covalent, bonds exclusively.

8. A semiconductor device formed of material substan-tially comprising throughout, a local order defined by groups of seven or more atoms covalently bonded together into substantially pentagonal tubes.

9. A semiconductor device formed of material substan-tially comprising throughout, a local order defined by poly-phosphide groups having the formula MPx where M is a metal and x is greater than six.

10. A semiconductor device substantially comprising throughout, a material having a local order defined by atoms joined together by multiple covalent atom-to-atom bonds forming layers of pentagonal tubes, the tubes within each layer being substantially parallel to each other.

11. A semiconductor device comprising an inorganic material having as its dominant local order substantially throughout its extent catenations of homatomic covalently bonded atoms, substantially all of the covalent bonds of said material being involved in the catenation of said atoms and said covalent bonds providing the dominant conduction paths in said material, said material having a bandgap sub-stantially within the range of 1.4 to 2.2eV.

12. A semiconductor device comprising an inorganic ma-terial having as its dominant local order substantially throughout its extent catenations of homatomic covalently bonded atoms, substantially all of the covalent bonds of said material being involved in the catenation of said atoms and said covalent bonds providing the dominant conduction paths in said material, said material having a photoconduc-tivity ratio substantially within the range of 100 to 10,000.

13. A semiconductor device comprising a high phosphor-us polyphosphide material condensed from the vapor phase in the presence of an alkali metal.

14. The invention defined in Claim 13 wherein said material is RbPx and x is at least 7.

15. The invention defined in claims 1, 2 or 3 wherein said material is CsPx and x is at least 7.

16. The invention defined in claims 7, 8 or 9 wherein said material is CsPx and x is at least 7.

17. The invention defined in claims 10, 11 or 12 wherein said material is CsPx and x is at least 7.

18. The invention defined in claim 13 wherein said material is CsPx and x is at least 7.

19. The method of chemical vapor deposition of high phosphorus polyphosphide materials comprising flowing separate gas streams of vaporized phosphorus and an alkali metal over a condensing substrate.

20. The method of flash evaporation comprising flowing a material through a heated susceptor to vaporize it.

21. A material produced by ball milling together one or more alkali metals and one or more pnictides.

22. The method of preparing polyphosphide materials of the formula MPx where M is one or more metals, x is at least 7 and P is mostly phosphorus but may include one or more pnictides by vapor transport comprising providing two heated separated sources of phosphorus and one or more metals and a separate deposition zone having a substantially constant temperature over an extended area.

23. The method of forming substantially pure phosphor-us by depositing the phosphorus by vapor transport on a sub-strate of a metal polyphosphide.

24. A method of making metal polyphosphide materials comprising heating one or more metals and phosphorus in a controlled temperature zone and depositing from vapor there-by supplied metal phosphide at a second controlled tempera-ture zone having a substantially constant temperature over an extended area.

25. The method of forming a high phosphorus poly-phosphide material having the formula MPx wherein M is one or more metals, x is equal to or greater than 15 and P is phos-phorus, comprising heating together M and P, and then cooling them to room temperature.

26. Crystalline phosphorus in the form of twisted fiber.

27. A crystalline polyphosphide material in the form of star shaped rods.

28. Monoclinic phosphorus material condensed from the vapor phase in the presence of a substantial amount of alkali metal vapor.

29. High purity phosphorus condensed from the vapor phase at a temperature substantially within the range of 500-550°C.

30. A material comprising a crystal of monoclinic phosphorus having a largest dimension substantially greater than 0.2mm.

31. A material comprising a crystal of monoclinic phosphorus having a smallest dimension substantially greater than 0.05mm.

32. An optical rotator formed of monoclinic phos-phorus material.

33. The method of preparing high purity phosphorus material comprising vaporizing phosphorus in a sealed ampoule and condensing it at a temperature substantially within the range of 500-550°C.

34. A composite material comprising monoclinic phosphorus.

35. A fire retardant material comprising a high phos-phorus polyphosphide.

36. A filler material comprising a high phosphorus polyphosphide.

37. A reinforcing material for composite materials comprising a crystalline high phosphorus polyphosphide.

38. A protective coating material comprising a high pnictide polypnictide.

39. An optical coating material comprising a high pnictide polypnictide.

40. An antireflection coating material comprising at least one layer of a high pnictide polypnictide.

41. The process of doping of high phosphorus semi-conductors with other pnictides.

42. The process of doping of high phosphorus semi-conductors with a metal having occupied d or f outer electronic levels.

43. A method of forming a semiconductor device com-prising the steps of:
a) providing a material which comprises, at least as one component thereof, a polyphosphide containing phosphorus-to-phosphorus bonds and containing alkali metal atoms which atoms are bonded to said phosphorus atoms, and wherein the number of con-secutive covalent phosphorus-to-phosphorus bonds is sufficiently greater than the number of non-phosphorus-to-phosphorus bonds, to render said material semiconducting; and b) attaching to said material means for electrically communicating with said material to utilize said material as a semiconductor.

44. A method of forming a semiconductor device com-prising the steps of:
a) providing a material which comprises, at least as one component thereof, at least two polyphosphide units, each unit having a skeleton of at least 7 covalently bonded phosphorus atoms, said units having associated therewith at least one alkali metal atom, said alkali metal atoms conductively bridging the phosphorus skeleton of one unit with the phosphorus skeleton of another unit, and said material having a band gap primarily determined by said phosphorus-to-phosphorus bonds; and b) attaching to said material means for electrically communicating with said material to utilize said material as a semiconductor.

45. A method of forming a semiconductor device comprising the steps of:
a) providing a material which comprises, at least as one component thereof, a polyphosphide having the formula MPx wherein M is an alkali metal and x is the atom ratio of P to M, x being at least 7, and wherein said material has an energy band gap with-in a range of from 1 to 3 eV; and b) attaching to said material means for electrically communicating with said material to utilize said material as a semiconductor.

46. A material produced by the method defined in any of Claims 18, 22 or 25.

47. A semiconductor device formed of the material defined in any of Claims 1, 2 or 3.

48. The invention defined in Claims 1, 2 or 3 wherein said material comprises a single alkali metal.

49. The invention defined in Claims 1, 2 or 3 wherein said material comprises at least two different alkali metals.

50. The invention defined in Claims 9, 43 or 45 wherein in said material at least 7 phosphorus atoms are bonded to other phosphorus atoms per each of a metal atom in said material.

51. The invention defined in Claims 9, 43 or 45 wherein 15 phosphorus atoms are bonded to other phosphorus atoms per each of a metal atom in said material.

52. The invention defined in Claims 9, 43 or 45 wherein there are at least 500 phosphorus atoms per each of a metal atom in said material.

53. The invention defined in Claims 9,43 or 45 wherein said material may be defined by the formula [MP7]a [P8]b, wherein b:a is the atom ratio of [P8] to [MP7].

54. The invention defined in Claims 1, 2 or 3 wherein said polyphosphide has the formula MPx and x is substantially equal to 15.

55. The invention defined in Claims 22, 24 or 27 wherein said polyphosphide has the formula MPx and x is substantially equal to 15.

56. The invention defined in Claims 28, 43 ox 44 wherein said polyphosphide has the formula MPx and x is substantially equal to 15.

57. The invention defined in Claim 45 wherein said poly-phosphide has the formula MPx and x is substantially equal to 15.

58. The invention defined in Claims 1, 2 or 3 wherein said polyphosphide has the formula MPx and x is greater than 14.

59. The invention defined in Claims 22, 24 or 27 wherein said polyphosphide has the formula MPx and x is greater than 14.

60. The invention defined in Claims 28, 43 or 44 wherein said polyphosphide has the formula MPx and x is greater than 14.

61. The invention defined in Claim 45 wherein said poly-phosphide has the formula MPx and x is greater than 14.

62. The invention defined in Claims 1, 2 or 3 wherein x is within a range of 7 to 15.

63. The invention defined in Claims 22, 24 or 27 wherein x is within a range of 7 to 15.

64. The invention defined in Claims 28, 43 or 44 wherein x is within a range of 7 to 15.

65. The invention defined in Claim 45 wherein x is within a range of 7 to 15.

66. The invention defined in Claims 1, 2 or 3 wherein said metal is Li.

67. The invention defined in Claims 22, 24 or 27 wherein said metal is Li.

68. The invention defined in Claims 28, 43 or 44 wherein said metal is Li.

69. The invention defined in Claim 45 wherein said metal is Li.

70. The invention defined in Claims 1, 2 or 3 wherein said metal is Na.

71. The invention defined in Claims 22, 24 or 27 wherein said metal is Na.

72. The invention defined in Claims 28, 43 or 44 wherein said metal is Na.

73. The invention defined in Claim 45 wherein said metal is Na.

74. The invention defined in Claims 1, 2 or 3 wherein said metal is K.

75. The invention defined in Claims 22, 24 or 27 wherein said metal is K.

76. The invention defined in Claims 28, 43 or 44 wherein said metal is K.

77. The invention defined in Claim 45 wherein said metal is K.

78. The invention defined in Claims 1, 2 or 3 wherein said metal is Rb.

79. The invention defined in Claims 22, 24 or 27 wherein said metal is Rb.

80. The invention defined in Claims 28, 43 or 44 wherein said metal is Rb.

81. The invention defined in Claim 45 wherein said metal is Rb.

82. The invention defined in Claims 1, 2 or 3 wherein said metal is Cs.

83. The invention defined in Claims 22, 24 or 27 wherein said metal is Cs.

84. The invention defined in Claims 28, 43 or 44 wherein said metal is Cs.

85. The invention defined in Claim 45 wherein said metal is Cs.

86. The invention defined in Claims 2, 7 or 8 wherein at least 6/7ths of said atoms have 3 homatomic bonds exclusively.

87. The invention defined in Claims 9, 10 or 11 wherein at least 6/7ths of said atoms have 3 homatomic bonds exclusively.

88. The invention defined in Claim 12 wherein at least 6/7ths of said atoms have 3 homatomic bonds exclusively.

89. The invention defined in Claims 2, 7 or 8 wherein at least 14/15ths of said atoms have 3 homatomic bonds.

90. The invention defined in Claims 9, 10 or 11 wherein at least 14/15ths of said atoms have 3 homatomic bonds.

91. The invention defined in Claim 12 wherein at least 14/15ths of said atoms have 3 homatomic bonds.

92. The invention defined in Claims 2, 7 or 8 wherein the proportion of atoms of said species having 3 homatomic bonds is very much greater than 14/15ths.

93. The invention defined in Claims 9, 10 or 11 wherein the proportion of atoms of said species having 3 homatomic bonds is very much greater than 14/15ths.

94. The invention defined in Claim 12 wherein the pro-portion of atoms of said species having 3 homatomic bonds is very much greater than 14/15ths.

95. The invention defined in Claims 2, 13 or 22 wherein said material is characterized in its local order by tubular columnar structure.

96. The invention defined in Claims 25, 43 or 44 wherein said material is characterized in its local order by tubular columnar structure.

97. The invention defined in Claim 45 wherein said material is characterized in its local order by tubular columnar structure.

98. The invention defined in Claims 2, 13 or 22 wherein said material is characterized in its local order by tubular columnar structure and said tubular structures in a local order are all generally parallel.

99. The invention defined in Claims 25, 43 or 44 wherein said material is characterized in its local order by tubular columnar structure and said tubular structures in a local order are all generally parallel.

100. The invention defined in Claim 45 wherein said mat-erial is characterized in its local order by tubular columnar structure and said tubular structures in a local order are all generally parallel.

101. The invention defined in Claims 2, 13 or 22 wherein the major component atoms of said material are trivalent.

102. The invention defined in Claims 25, 43 or 44 wherein the major component atoms of said material are trivalent.

103. The invention defined in Claim 45 wherein the major component atoms of said material are trivalent.

104. The invention defined in Claims 2, 13 or 22 wherein said material is characterized in its local order by tubular columnar structure and said tubular structures in a local order are all generally parallel said columnar structure is channel-like when viewed on end.

105. The invention defined in Claims 25, 43 or 44 wherein said material is characterized in its local order by tubular columnar structures and said tubular structures in a local order are all generally parallel said columnar structure is channel-like when viewed on end.

106. The invention defined in Claim 45 wherein said material is characterized in its local order by tubular columnar structures and said tubular structures in a local order are all generally parallel said columnar structure is channel-like when viewed on end.

107. The invention defined in Claims 2, 7 or 8 wherein said material is characterized in its local order by tubular columnar structures and said tubular structures in a local order are all generally parallel said columnar structure is pentagonal when viewed on end.

108. The invention defined in Claims 25, 43 or 44 wherein said material is characterized in its local order by tubular columnar structures and said tubular structures in a local order are all generally parallel, said columnar structure is pentagonal when viewed on end.

109. The invention defined in Claim 45 wherein said material is characterized in its local order by tubular columnar structures and said tubular structures in a local order are all generally parallel, said columnar structure is pentagonal when viewed on end.

110. The invention defined in Claims 2, 7 or 8 wherein said atoms are one or more pnictides.

111. The invention defined in Claims 25, 43 or 44 wherein said atoms are one or more pnictides.

112. The invention defined in Claim 45 wherein said atoms are one or more pnictides.

113. The invention defined in Claims 2, 7 or 8 wherein said atoms are predominantly phosphorus.

114. The invention defined in Claims 25, 43 or 44 wherein said atoms are predominantly phosphorus.

115. The invention defined in Claim 45 wherein said atoms are predominantly phosphorus.

116. A semiconductor device as defined in Claims 7, 8 or 11 wherein the bonds of said atoms are spaced at an average angle greater than 98°.

117. A semiconductor device as defined in Claims 7, 8 or 11 wherein the bonds of said atoms are spaced at an average angle in the range of from 87° to 109°.

118. A semiconductor device as defined in Claim 11 and additional atoms of one or more different elements than atoms of said catenations bonded between two or more of said cat-enations.

119. A semiconductor device as defined in Claim 12 and additional atoms of one or more different elements than atoms of said catenations bonded between two or more of said cat-enations.

120. A semiconductor device as defined in Claim 118 wherein said additional atoms form conduction paths between the caten-ations to which they are bonded.

121. A semiconductor device as defined in Claims 119 wherein said additional atoms form conduction paths between the catenations to which they are bonded.

122. A semiconductor device as defined in Claims 118 or 119 wherein said catenations in each local order are all generally parallel.

123. A semiconductor device as defined in Claims 120 or 121 wherein said catenations in each local order are all generally parallel.

124. The invention defined in Claims 1, 2 or 3 wherein said material is formed as the deposition product from vapor transport in a deposition zone from separated sources of phos-phorus and an alkali metal.

125. The invention defined in Claims 22, 24 or 25 wherein said material is formed as the deposition product from vapor transport in a deposition zone from separated sources of phos-phorus and an alkali metal.

126. The invention defined in Claims 43, 44 or 45 wherein said material is formed as the deposition product from vapor transport in a deposition zone from separated sources of phos-phorus and an alkali metal.

127. The invention defined in Claims 7, 8 or 10 wherein the substantial majority of said atoms are phosphorus.

128. The invention defined in Claims 43, 44 or 45 wherein the substantial majority of said atoms are phosphorus.

129. The invention defined in Claims 3, 9 or 30 wherein the atom ratio of phosphorus to metal is substantially 50 or greater.

130. The invention defined in Claims 3, 9 or 30 wherein the atom ratio of phosphorus to metal is substantially 200 or greater.

131. The invention defined in Claims 3, 9 or 30 wherein the atom ratio of phosphorus to metal is substantially 1,000 or greater.

132. The invention defined in Claims 3, 9 or 30 wherein the atom ratio of phosphorus to metal is substantially 1000 or greater and the amount of said metal is less than 1,000 parts per million.

133. The invention defined in Claims 1, 2 or 3 wherein said material has a bandgap substantially within the range of 1 to 3 electron volts.

134. The invention defined in Claims 7, 8 or 9 wherein said material has a bandgap substantially within the range of 1 to 3 electron volts.

135. The invention defined in Claims 10, 11 or 12 wherein said material has a bandgap substantially within the range of 1 to 3 electron volts.

136. The invention defined in Claim 13 wherein said material has a bandgap substantially within the range of 1 to 3 electron volts.

137. The invention defined in Claims 1, 2 or 3 wherein said material has a bandgap substantially within the range of 1.4 to 2.2 electron volts.

138. The invention defined in Claims 7, 8 or 9 wherein said material has a bandgap substantially within the range of 1.4 to 2.2 electron volts.

139. The invention defined in Claims 10, 11 or 12 wherein said material has a bandgap substantially within the range of 1.4 to 2.2 electron volts.

140. The invention defined in Claim 13 wherein said material has a bandgap substantially within the range of 1.4 to 2.2 electron volts.

141. The invention defined in Claims 1, 2 or 3 wherein said material has a bandgap of substantially 1.8 electron volts.

142. The invention defined in Claims 7, 8 or 9 wherein said material has a bandgap of substantially 1.8 electron volts.

143. The invention defined in Claims 10, 11 or 12 wherein said material has a bandgap of substantially 1.8 electron volts.

144. The invention defined in Claim 13 wherein said material has a bandgap of substantially 1.8 electron volts.

145. The invention defined in any of Claims 1, 2 or 3 wherein said material has a photoconductivity ratio within the range of 100 to 10,000.

146. The invention defined in any of Claims 7, 8 or 9 wherein said material has a photoconductivity ratio within the range of 100 to 10,000.

147. The invention defined in any of Claims 10, 11 or 12 wherein said material has a photoconductivity ratio within the range of 100 to 10,000.

148. The invention defined in Claim 13 wherein said material has a photoconductivity ratio within the range of 100 to 10,000.

149. The invention defined in any of Claims 1, 2 or 3 wherein said material is formed of a single crystal.

150. The invention defined in any of Claims 7, 8 or 9 wherein said material is formed of a single crystal.

151. The invention defined in any of Claims 10, 11 or 12 wherein said material is formed of a single crystal

152. The invention defined in Claim 13 wherein said material is formed of a single crystal.

153. The invention defined in any of Claims 1, 2 or 3 wherein said material is polycrystalline.

154. The invention defined in any of Claims 7, 8 or 9 wherein said material is polycrystalline.

155. The invention defined in any of Claims 10, 11 or 12 wherein the said material is polycrystalline.

156. The invention defined in Claim 13 wherein said material is polycrystalline.

157. The invention defined in any of Claims 1, 2 or 3 wherein said material is amorphous.

158. The invention defined in any of Claims 7, 8 or 9 wherein said material is amorphous.

159. The invention defined in any of Claims 10, 11 or 12 wherein said material is amorphous.

160. The invention defined in Claim 13 wherein said mat-erial is amorphous.

161. The invention defined in Claims 1, 2 or 3 wherein said material is in the form of a thin film.

162. The invention defined in Claims 7, 8 or 9 wherein said material is in the form of a thin film.

163. The invention defined in Claims 10, 11 or 12 wherein said material is in the form of a thin film.

164. The invention defined in Claim 13 wherein said material is in the form of a thin film.

165. The invention defined in Claims 1, 2 or 3 wherein said material is in a form of a thin film and is deposited on a glass substrate.

166. The invention defined in claims 7, 8 or 9 wherein said material is in a form of a thin film and is deposited on a glass substrate.

167. The invention defined in Claims 10, 11 or 12 wherein said material is in a form of a thin film and is deposited on a glass substrate.

168. The invention defined in Claim 13 wherein said material is in a form of a thin film and is deposited on a glass substrate.

169. The invention defined in Claims 1, 2 or 3 wherein said material is in a form of a thin film and is deposited on a metal substrate.

170. The invention defined in Claims 7, 8 or 9 wherein said material is in a form of a thin film and is deposited on a metal substrate.

171. The invention defined in Claims 10, 11 or 12 wherein said material is in a form of a thin film and is deposited on a metal substrate.

172. The invention defined in Claim 13 wherein said material is in a form of a thin film and is deposited on a metal substrate.

173. A semiconductor device as defined in Claims 7, 8 or 9 further defined as comprising a junction.

174. A semiconductor device as defined in Claims 10, 11 or 12 further defined as comprising a junction.

175. A semiconductor device as defined in Claim 13 fur-ther defined as comprising a junction.

176. A semiconductor device as defined in Claims 7, 8 or 9 in the form of a junction wherein said junction comprises a metal selected from the group consisting of Cu, Al, Mg, Ni, Au, Ag, and Ti.

177. A semiconductor device as defined in Claims 10, 11 or 12 in the form of a junction wherein said junction comprises a metal selected from the group consisting of Cu, Al, Mg, Ni, Au, Ag, and Ti.

178. A semiconductor device as defined in Claim 13 in the form of a junction wherein said junction comprises a metal selected from the group consisting of Cu, Al, Mg, Ni, Au, Ag, and Ti.

179. A semiconductor device as defined in Claims 7, 8 or 9 in the form of a metal junction wherein said junction metal is Ni.

180. A semiconductor device as defined in Claims 10, 11 or 12 in the form of a metal junction wherein said junction metal is Ni.

181. A semiconductor device as defined in Claim 13 in the form of a metal junction wherein said junction metal is Ni.

182. A semiconductor device as defined in Claims 7, 8 or 9 wherein said material is doped with atoms of another pnictide.

183. A semiconductor device as defined in Claims 10, 11 or 12 wherein said material is doped with atoms of another pnictide.

184. A semiconductor device as defined in Claim 13 wherein said material is doped with atoms of another pnictide.

185. A semiconductor device as defined in Claims 7, 8 or 9 wherein said material is doped with atoms of another pnic-tide which is As.

186. A semiconductor device as defined in Claims 10, 11 or 12 wherein said material is doped with atoms of another pnic-tide which is As.

187. A semiconductor device as defined in Claim 13 wherein said material is doped with atoms of another pnictide which is As.

188. A semiconductor device as defined in Claims 7, 8 or 9 wherein said material is doped by diffusing therein a metal having occupied outer f or d electronic levels.

189. A semiconductor device as defined in Claims 10, 11 or 12 wherein said material is doped by diffusing therein a metal having occupied outer f or d electronic levels.

190. A semiconductor device as defined in Claim 13 wherein said material is doped by diffusing therein a metal having occupied outer f or d electronic levels.

191. A semiconductor device as defined in Claims 7, 8 or 9 wherein said material is doped by diffusing therein a dopant chosen from the group consisting of nickel, iron and chromium.

192. A semiconductor device as defined in Claims 10, 11 or 12 wherein said material is doped by diffusing therein a dopant chosen from the group consisting of nickel, iron and chromium.

193. A semiconductor device as defined in Claim 13 wherein said material is doped by diffusing therein a dopant chosen from the group consisting of nickel, iron and chromium.

194. A semiconductor device as defined in Claims 7, 8 or 9 comprising a metal contact and said contact being selected from the group comprising Cu, Al, Mg, Ni, Au, As, and Ti.

195. A semiconductor device as defined in Claims 10, 11 or 12 comprising a metal contact and said contact being selected from the group comprising Cu, Al, Mg, Ni, Au, As, and Ti.

196. A semiconductor device as defined in Claim 13 com-prising a metal contact and said contact being selected from the group comprising Cu, Al, Mg, Ni, Au, As and Ti.

197. The doping defined in Claim 41 wherein said pnictides are selected from the group consisting of As, Sb, and Bi.

198. The doping defined in Claim 196 wherein said pnic-tide is As.

199. The doping defined in Claim 42 wherein said metal is selected from the group consisting of Fe, Ni, or Cr.

200. The invention defined in Claims 1, 2 or 3 wherein said material comprises an alkali metal.

201. The invention defined in Claims 7, 8 or 9 wherein said material comprises an alkali metal.

202. The invention defined in Claims 10, 11 or 12 wherein said material comprises an alkali metal.

203. The invention defined in Claim 13 wherein said mat-erial comprises an alkali metal.

204. The invention defined in Claims 1, 2 or 3 wherein said material comprises an alkali metal and at least 7 phos-phorus atoms are bonded to other phosphorus atoms per each of a metal atoms in said material.

205. The invention defined in Claims 7, 8 or 9 wherein said material comprises an alkali metal and at least 7 phos-phorus atoms are bonded to other phosphorus atoms per each of a metal atoms in said material.

206. The invention defined in Claims 10, 11 or 12 wherein said material comprises an alkali metal and at least 7 phos-phorus atoms are bonded to other phosphorus atoms per each of a metal atoms in said material.

207. The invention defined in Claim 13 wherein said material comprises an alkali metal and at least 7 phosphorus atoms are bonded to other phosphorus atoms per each of a metal atoms in said material.

208. The invention defined in Claims 1, 3 or 13 wherein said metal is Li.

209. The invention defined in Claims 1, 3 or 13 wherein said metal is Na.

210. The invention defined in Claims 1, 3 or 13 wherein said metal is K.

211. The invention defined in Claims 1, 3 or 13 wherein said metal is Rb.

212. The invention defined in Claims 1, 3 or 13 wherein said metal is Cs.

213. The invention defined in claims 1, 2 or 3 wherein said material is LiPx and x is at least 7.

214. The invention defined in Claims 7, 8 or 9 wherein said material is LiPx and x is at least 7.

215. The invention defined in Claims 10, 11 or 12 wherein said material is LiPx and x is at least 7.

216. The invention defined in Claim 13 wherein said material is LiPx and x is at least 7.

217. The invention defined in Claims 1, 2 or 3 wherein said material is NaPx and x is at least 7.

218. The invention defined in Claims 7, 8 or 9 wherein said material is NaPx and x is at least 7.

219. The invention defined in Claims 10, 11 or 12 wherein said material is NaPx and x is at least 7.

220. The invention defined in Claim 13 wherein said material is NaPx and x is at least 7.

221. The invention defined in Claims 1, 2 or 3 wherein said material is KPx and x is at least 7.

222. The invention defined in Claims 7, 8 or 9 wherein said material is KPx and x is at least 7.

223. The invention defined in Claims 10, 11 or 12 wherein said material is KPx and x is at least 7.

224. The invention defined in Claim 13 wherein said material is KPx and x is at least 7.

225. The invention defined in Claims 1, 2 or 3 wherein said material is RbPx and x is at least 7.

226. The invention defined in Claims 7, 8 or 9 wherein said material is RbPx and x is at least 7.

227. The invention defined in Claims 10, 11 or 12 wherein said material is RbPx and x is at least 7.

228. High purity phosphorus as defined in Claim 29 wherein said high purity phosphorus is condensed in the presence of a substantial amount of alkali metal vapor.

229. High purity phophorus as defined in Claim 29 wherein said high purity phosphorus is condensed at a temperature sub-stantially equal to 539°C in the presence of a substantial amount of alkali metal vapor.

230. Monoclinic phosphorus as defined in Claims 30 or 31, condensed from the vapor phase in the presence of a substantial amount of alkali metal vapor.

231. Monoclinic phosphorus as defined in Claims 3 or 31 condensed from the vapor phase at a temperature substantially within the range of 500-550°C in the presence of a substantial amount of alkali metal vapor.

232. The method defined in Claim 20 wherein the flowing material comprises phosphorus.

233. The method defined in Claim 20 wherein the flowing material comprises an alkali metal.

234. Monoclinic phosphorus as defined in Claims 28 or 30 wherein said phosphorus contains alkali metal substantially within the range of 50-2000 parts per million.

235. Monoclinic phosphorus as defined in Claims 28 or 30 in a platelet like habit.

236. Monoclinic phosphorus as defined in Claims 28 or 30 in a truncated pyramid habit.

237. Monoclinic phosphorus as defined in Claims 28 or 30 wherein said vapor is formed at a temperature within the range of 546-564°C.

238. The invention defined in Claim 28 wherein said alkali metal is sodium.

239. The invention defined in Claim 28 wherein said alkali metal is potassium.

240. The invention defined in Claim 28 wherein said alkali metal is rubidium.

241. The invention defined in Claim 28 wherein said alkali metal is cesium.

242. Phosphorus as defined in Claim 26, condensed from the vapor phase at substantially 509°C.

243. A polyphosphide as defined in Claim 27, condensed from the vapor phase at a temperature below 500°C.

244. The polyphosphide defined in Claim 27 condensed at substantially 462°C.

245. A coating as defined in Claims 38, 39 or 40 on a glass substrate.

246. A coating as defined in Claims 38, 39 or 40 on a metal substrate.

247. A coating as defined in Claims 38, 39 or 40 wherein said pnictide is phosphorus.

248. A coating as defined in Claims 38, 39 or 40 wherein said coating is amorphous.

249. A reinforcing material for composite materials com-prising a crystalline MPx where M comprises an alkali metal, P is phosphorus, and x is equal to or greater than 7.

250. A reinforcing material for composite materials com-prising a crystalline MP15 where M comprises an alkali metal and P is phosphorus.

251 . A reinforcing material for composite materials comprising a crystalline MPx where M comprises an alkali metal, or other metal or metals mimicking the bonding of an alkali metal, P is phosphorus, and x is equal to or greater than 7.

252. A reinforcing material for composite materials comprising a crystalline MP15 where M comprises an alkali metal, or other metal or metals mimicking the bonding of an alkali metal, and P is phosphorus.