Defects in the crystal lattice of diamond are common. They may be the result of extrinsic substitutional or interstitial impurities or intrinsic lattice defects. All diamonds possess lattice defects of some sort; the defects themselves may be produced during or after the diamond growth. The material properties of diamond are affected by these defects and determine to which type a diamond is assigned; the most dramatic effects are on the diamond color and electrical conductivity, as explained by the band theory.
The defects can be detected by different types of spectroscopy, including electron paramagnetic resonance (EPR), luminescence induced by light (photoluminescence, PL) or electron beam (cathodoluminescence, CL), and absorption of light in the infrared (IR), visible and UV parts of the spectrum. The resulting absorption spectrum can then be analyzed, identified, and even used to separate natural from synthetic or enhanced diamonds.
The number of defects in diamond whose microscopic structure has been reliably identified is rather large (many dozens), and only the major ones are briefly mentioned in this article.
Extrinsic defects
Various elemental analyses of diamond reveal a wide range of impurities. They however mostly originate from inclusions of foreign materials in diamond, which could be nanometer-small and invisible in an optical microscope. Also, virtually any element can be hammered into diamond by ion implantation. More essential however are elements which can be introduced into the diamond lattice as isolated atoms (or small atomic clusters) during the diamond growth. By 2008, those elements are nitrogen, boron, hydrogen, silicon, phosphorus, nickel, cobalt and maybe sulfur. Elements manganese and tungsten and maybe beryllium have been unambiguously detected in diamond, but they could originate from foreign inclusions. Detection of isolated iron in diamond has been later reassigned to micro-particles of ruby, produced during the diamond synthesis. Oxygen is believed to be a major impurity in diamond, but it has not been spectroscopically identified in diamond yet. Two electron paramagnetic resonance centers (OK1 and N3 - not to be confused with the optical N3 center) have been assigned to nitrogen-oxygen complexes. However, the assignment is indirect and the corresponding concentrations are rather low (parts per million).
Nitrogen
The most common impurity in diamond is nitrogen, which can comprise up to 1% of a diamond by mass. Previously, all lattice defects in diamond were thought to be the result of structural anomalies; later research revealed nitrogen to be present in most diamonds and in many different configurations. It is believed that nitrogen enters the diamond lattice as a single atom (i.e. nitrogen-containing molecules dissociate before incorporation), however it was proven that molecular nitrogen incorporates into diamond as well.
Absorption of light and other material properties of diamond are highly dependent upon nitrogen content and aggregation state. Although all aggregate configurations cause absorption in the infrared, diamonds containing aggregated nitrogen are usually colorless, i.e. have little absorption in the visible spectrum. The four main nitrogen forms are as follows:
C-nitrogen center
C center corresponds to electrically neutral single substitutional nitrogen atoms in the diamond lattice. These are easily seen in electron paramagnetic resonance spectra (in which they are confusingly called P1 centers - labeling of diamond defects is historical and not systematic). C centers impart a deep yellow to brown color; these diamonds are classed type Ib and are commonly known as "canary diamonds", which are rare in gem form. Most synthetic diamonds produced by high-pressure high-temperature (HPHT) technique contain a high level of nitrogen in the C form; this nitrogen impurity originates from the atmosphere or from the graphite source. One nitrogen atoms per 100,000 carbon atoms will produce yellow color. Because the nitrogen atoms have five available electrons (one more than the carbon atoms they replace), they act as "deep donors"; that is, each substituting nitrogen has an extra electron to donate and forms a donor energy level within the band gap. Light with energy above ca. 2.2 eV and above can excite the donor electrons into the conduction band, thereby allowing light absorption.
The C center produces a characteristic infrared absorption spectrum with a sharp peak at 1344 cm-1 and a broader peak at 1130cm-1. Absorption at those peaks is routinely used to measure the concentration of single nitrogen. Another proposed way, using the UV absorption at ~260 nm, has been later discarded as unreliable. Acceptor defects in diamond ionize the fifth nitrogen electron in the C center converting it into C+ center. The latter has a characteristic IR absorption spectrum with a sharp peak at 1332 cm-1 and broader and weaker peaks at 1115, 1046 and 950 cm-1.
A-nitrogen center
The A center is probably the most common defect in natural diamonds. It consists of a neutral nearest-neighbor pair of nitrogen atoms substituting for the carbon atoms. The A center produces UV absorption threshold at ~4 eV (310 nm, i.e. invisible to eye) and thus causes no coloration; these diamonds are classed as type IaA.
The A center is diamagnetic, but if ionized by UV light or deep acceptors, it produces an electron paramagnetic resonance spectrum W24, whose analysis unambiguously proves the N=N structure.
The A center shows an IR absorption spectrum with no sharp features, which is distinctly different from that of the C or B centers. Its strongest peak at 1282 cm-1 is routinely used to estimate the nitrogen concentration in the A form.
B-nitrogen center
There is a general consensus that B center (sometimes called B1) consists of a carbon vacancy surrounded by four nitrogen atoms substituting for carbon atoms. This model is consistent with other experimental results, but there is no any direct spectroscopic data corroborating it. Diamonds where most nitrogen forms B centers are rare and are classed as type IaB; most gem diamonds contain a mixture of A and B centers, together with N3 centers, the combination producing the yellow-brown Cape series.
Similar to the A centers, B centers do not induce color. In fact, no UV or visible absorption can be attributed to the B center. Early assignment of the N9 absorption system to the B center have been disproven later. The B center has a characteristic IR absorption spectrum (see the infrared absorption picture above) with a sharp peak at 1332 cm-1 and a broader peak at 1280cm-1. The latter is routinely used to estimate the nitrogen concentration in the B form.
Note that many optical peaks in diamond accidentally have similar spectral positions, which causes much confusion among gemologists. The silent agreement in scientific literature is to use for defect identification the whole spectrum rather than one peak, and to consider the history of the growth and processing of individual diamond.
N3 nitrogen center
The N3 center consists of three nitrogen atoms surrounding a vacancy. Its concentration is always just a fraction of the A and B centers. The N3 center is paramagnetic, so its structure is well justified from the analysis of the EPR spectrum P2. This defect produces a characteristic absorption and luminescence line at 415 nm and thus does not induce color on its own. However, N3 center is always accompanied by the N2 center, having an absorption line at 478 nm (and no luminescence). As a result, diamonds rich in N3/N2 centers are yellow in color.
Boron
Diamonds containing boron as a substitutional impurity are termed type IIb. Only one percent of diamonds are of this type, and most are blue to grey. The boron acts as an acceptor; that is, because the substituting boron atoms have one less available electron than the carbon atoms they replace, each boron atom creates an electron hole in the band gap that can accept an electron from the valence band. This allows red light absorption, and due to the small energy (0.37 eV) needed for the electron to leave the valence band, holes are created in the latter even via thermal heat at room temperatures. These holes can move in an electric field and render the diamond electrically conductive (i.e., a p-type semiconductor). Very little substitutional boron is required for this to happen—a typical ratio is one boron atom per 1,000,000 carbon atoms.
Type IIb diamonds transmit in the ultraviolet down to ca. 250 nm but do not absorb in the visible region apart from the far red (hence the blue color); they may phosphoresce blue after exposure to shortwave ultraviolet. Apart from optical absorption, boron acceptors have been detected by electron paramagnetic resonance.
Phosphorus
Phosphorus has been intentionally introduced into diamond grown by chemical vapor deposition (CVD) at concentrations up to ~0.01%. Phosphorus substitutes carbon in the diamond lattice. Similar to nitrogen, phosphorus has one more electron than carbon and thus acts as a donor; however, the ionization energy of phosphorus (0.6 eV) is much smaller than of nitrogen (1.7 eV) and is small enough for room-temperature electronic applications, such as p-n junctions and UV light emitting diodes (LEDs, at 235 nm).
Hydrogen
Hydrogen-related defects are very different in natural diamond and synthetic diamond films. Those films are produced by various chemical vapor deposition (CVD) techniques in an atmosphere rich in hydrogen (typical hydrogen/carbon ratio >100), under strong bombardment of growing diamond by the plasma ions. As a result, CVD diamond is always rich in hydrogen and vacancies. In polycrystalline films, or low-quality single crystal growth, much of the hydrogen may be located at the boundaries between diamond 'grains', or in non-diamond phase carbon inclusions. Within the diamond lattice itself, hydrogen-vacancy and hydrogen-nitrogen-vacancy complexes have been identified in negative charge states by electron paramagnetic resonance. In addition, numerous hydrogen-related IR absorption peaks are documented.
It is experimentally demonstrated that hydrogen passivates electrically active boron and phosphorus impurities. When hydrogen passivates boron acceptors, it possibly produces shallow donor centers.
In natural diamonds, several hydrogen-related IR absorption peaks have been observed, the strongest ones located at 1405, 3107 and 3237 cm-1 (see IR absorption figure above). However, the microscopic structure of the corresponding defects is yet unknown and it is even uncertain whether those defects originate in diamond or in foreign inclusions. Gray color in some diamonds from the Argyle mine in Australia is often associated with those hydrogen defects, but again, this assignment is yet unproven.
Nickel and Cobalt
When diamonds are grown by the high-pressure high-temperature technique, nickel, cobalt or some other metals are usually added into the growth medium to facilitate catalytically the conversion of graphite into diamond. As a result, metallic inclusions are often created. However, isolated nickel and cobalt atoms incorporate into diamond lattice, as demonstrated through characteristic hyperfine structure in electron paramagnetic resonance, optical absorption and photoluminescence spectra, and concentration of isolated nickel can reach 0.01%. This fact is by all means unusual considering the large difference in size between carbon and transition metal atoms and the superior rigidity of the diamond lattice.
A large number of Ni-related defects have been detected by electron paramagnetic resonance, optical absorption and photoluminescence. Three major structures can be distinguished: substitutional Ni, nickel-vacancy and nickel-vacancy decorated by one or more substitutional nitrogen atoms. The "nickel-vacancy" structure, also called "semi-divacancy" is rather special for most large impurities in diamond and silicon (e.g. tin in silicon). Their production mechanism is generally accepted as follows: large nickel atom, when incorporates substitutionally, expels a nearby carbon (i.e. creating a neighboring vacancy) and shifts in between the two sites.
Isolated Ni-related defects are routinely observed not only in synthetic, but also in natural diamonds.
Although the physical and chemical properties of cobalt and nickel are rather similar, the concentrations of isolated Co in diamond are much smaller than that of Ni. Several defects related to isolated cobalt have been detected by electron paramagnetic resonance and photoluminescence, but their structure is yet unknown.
An interesting phenomenon has been observed that when produced by the HPHT synthesis, low-symmetry Ni or Co defects have specific, non-averaged orientation in the diamond lattice determined by the direction of the growth planes.
Silicon
Silicon is a common impurity in diamond films grown by chemical vapor deposition and it originated either from Si substrate or silica windows or walls of the CVD reactor. Silicon has been detected in diamond lattice through sharp optical absorption at 738 nm and electron paramagnetic resonance. Similar to Ni, this form has been identified as Si-vacancy complex (semi-divacancy site) with a donor ionization energy of 2 eV. Si-vacancies however constitute minor fraction of total silicon. It is believed (though no proof exists) that much silicon substitutes for carbon thus becoming invisible for most spectroscopic techniques because Si and carbon atoms have the same configuration of the outer electronic shells.
Similar to nickel, isolated silicon is also detected not only in synthetic, but also in natural diamonds.
Sulfur
Around the year 2000, there was a wave of attempts to dope synthetic CVD diamond films by sulfur aiming at n-type conductivity with low activation energy. Successful reports have been published, but then dismissed as the conductivity was associated not with sulfur, but with residual boron, which is a highly efficient dopant in diamond.
As to 2008, there is only one reliable evidence (through hyperfine interaction structure in electron paramagnetic resonance) for isolated sulfur in diamond. The corresponding center called W31 has been observed in natural type-Ib diamonds in small concentrations (parts per million). It was assigned to a sulfur-vacancy complex - again, as in case of Ni and Si, a semi-divacancy site.
Intrinsic defects
The easiest way to produce intrinsic defects in diamond is by knocking off a carbon atom through irradiation with high-energy particles, such as alpha (helium), beta (electrons) or gamma particles, protons, neutrons, ions, etc. The irradiation can happen in the laboratory or in the nature (see Diamond enhancement - Irradiation); it produces primary defects named interstitials (knocked off carbon atoms) and remaining lattice vacancies. An important difference between the vacancies and interstitials in diamond is that whereas interstitials are mobile during the irradiation, even if performed at refrigerating temperatures, however vacancies start migrating only at temperatures ~700 0C.
It should be noted that vacancies and interstitials can also be produced in diamond by plastic deformation, though in much smaller concentrations.
Isolated carbon interstitial
Isolated interstitial has never been observed in diamond and is rendered unstable. Its interaction with a regular carbon lattice atom produces a "split-interstitial", a defect where two carbon atoms share a lattice site and are covalently bonded with the carbon neighbors. This defect has been thoroughly characterized by electron paramagnetic resonance (R2 center) and optical absorption, and unlike most other defects in diamond, it does not produce photoluminescence.
Interstitial complexes
The isolated split-interstitial moves during irradiation producing larger complexes of two and three split-interstitials, identified by electron paramagnetic resonance (R1 and O3 centers), optical absorption and photoluminescence.
Vacancy-Interstitial complexes
Most high-energy particles, beside knocking off a carbon atom from the lattice site, also pass it enough surplus energy for a rapid migration through the lattice. However, when relatively gentle gamma irradiation is used, this extra energy is minimal. The interstitials remain near the original vacancies producing vacancy-interstitials pairs identified through optical absorption.
Vacancy-di-interstitial pairs have been also produced, though by electron irradiation and through a different mechanism, and characterized by electron paramagnetic resonance.
Isolated vacancy
Isolated vacancy is perhaps the most studied defect in diamond, both experimentally and theoretically. Its most important practical property is optical absorption, like in the color centers, which gives diamond green, or sometimes even green-blue color (in pure diamond). The characteristic feature of this absorption is a series of sharp lines called GR1, 2, .. 8 (GR from "general radiation"), where GR1 line at 741 nm is the most prominent and important.
The vacancy behaves as a deep electron donor/acceptor, whose electronic properties depend on the charge state. The energy level for the +/0 states is at 0.6 eV and for the 0/- states is at 2.5 eV above the valence band.
Multivacancy complexes
Upon annealing of pure diamond at about 700 0C, vacancies migrate and form divacancies, characterized by optical absorption and electron paramagnetic resonance. Similar to single interstitials, divacancies do not produce photoluminescence. Divacancies, in turn, anneal out at ~900 0C creating multivacancy chains detected by EPR and presumably hexavacancy rings. The latter should be invisible to most spectroscopies, and indeed, they have not been detected so far. Annealing of vacancies change diamond color from green to yellow-brown. Similar mechanism (vacancy aggregation) is also believed to cause brown color of plastically deformed natural diamonds.
Dislocations
Dislocations are one of the most common structural defects in natural diamond. There are two important types of dislocations in diamond: the glide set, in which bonds break between layers of atoms with different indices (those not lying directly above each other); and the shuffle set, in which the breaks occur between atoms of the same index. The dislocations produce dangling bonds which introduce energy levels into the band gap, enabling the absorption of light. Broadband blue photoluminescence has been reliably identified with dislocations by direct observation in an electron microscope, however, it was noted that not all dislocations are luminescent and no clear correlation is known between the dislocation type and the parameters of the emission.
Platelets
Most natural diamonds contain extended planar defects in the <100> lattice planes, which are called platelets. Their size ranges from nanometers to many microns, and large ones are easily observed in an optical microscope. For a long time, platelets were tentatively associated with large nitrogen complexes - nitrogen sinks produced as a result of nitrogen aggregation at high temperature of the diamond synthesis. However, direct measurement of nitrogen in the platelets by EELS (an analytical technique of electron microscopy) revealed very little nitrogen. The currently accepted model of platelets is a large regular array of carbon interstitials.
Platelets produce sharp IR absorption peaks at 1359-1375 and 330 cm-1 in IR absorption spectra; where the position of the first peak depends on the platelet size. As with dislocations, a broad photoluminescence centered at about 1000 nm was associated with platelets by direct observation in an electron microscope. By studying this luminescence, it was deduced that platelets have a "bandgap" of ~1.7 eV.
Voidites
Voidites are octahedral nanometer-sized clusters present in many type-Ia diamonds, as revealed by electron microscopy. Laboratory experiments demonstrated that annealing of type-IaB diamond at high temperatures and pressures (>2600 0C) results in break-up of the platelets and formation of dislocation loops and voidites. Contrary to platelets, voidites do contain much nitrogen, in the molecular form.
Interaction between intrinsic and extrinsic defects
Extrinsic and intrinsic defects can interact producing new defect complexes. Such interaction usually occurs if a diamond containing extrinsic defects (impurities) is either plastically deformed or is irradiated and annealed.
Most important is interaction of vacancies and interstitials with nitrogen. Carbon interstitials react with substitutional nitrogen producing a bond-centered nitrogen interstitial showing strong IR absorption at 1450 cm-1. Vacancies are efficiently trapped by the A, B and C nitrogen centers. The trapping rate is the highest for the C centers, 8 times lower for the A centers and 30 times lower for the B centers. The C center (single nitrogen) by trapping a vacancy forms a famous nitrogen-vacancy center. The defect has been shown to exist in either a neutral or negative charge state and the negatively charged version may have important applications in quantum computing. A and B centers upon trapping a vacancy create corresponding 2N-V (H3 and H2 centers - where H2 is simply a negatively charged H3 center) and the neutral 4N-2V (H4 center).
Boron interacts with carbon interstitials forming a neutral boron–interstitial complex with a sharp optical absorption at 0.552 eV (2250 nm). No evidence is known for complexes of boron and vacancy. In contrast silicon does react with vacancies, creating the described above optical absorption at 738 nm. The assumed mechanism is trapping of migrating vacancy by substitutional Si resulting in the Si-V (semi-divacancy) configuration.
Similar mechanism is expected for Ni, for which both substitutional and semi-divacancy configurations are reliably identified (see subsection "nickel and cobalt" above). In an unpublished study, diamonds rich in substitutional Ni were electron irradiated and annealed with careful optical measurements performed at each annealing step, but no reliable evidence for creation or enhancement of Ni-vacancy centers could be obtained.
No comments:
Post a Comment