|Home | About | Journals | Submit | Contact Us | Français|
The II–V group semiconductors, with narrow band gaps, are important materials with many applications in infrared detectors, lasers, solar cells, ultrasonic multipliers, and Hall generators. Since the first report on trumpet-like Zn3P2nanowires, one-dimensional (1-D) nanostructures of II–V group semiconductors have attracted great research attention recently because these special 1-D nanostructures may find applications in fabricating new electronic and optoelectronic nanoscale devices. This article covers the 1-D II–V semiconducting nanostructures that have been synthesized till now, focusing on nanotubes, nanowires, nanobelts, and special nanostructures like heterostructured nanowires. Novel electronic and optoelectronic devices built on 1-D II–V semiconducting nanostructures will also be discussed, which include metal–insulator-semiconductor field-effect transistors, metal-semiconductor field-effect transistors, andp–nheterojunction photodiode. We intent to provide the readers a brief account of these exciting research activities.
Because of the ability to synthesize them in numerous configurations and their unique physical, chemical, optical, electrical, and magnetic properties, one-dimensional (1-D) nanostructures have attracted great interest in recent years [1-5]. 1-D nanostructures play important roles both as interconnect and functional units in fabricating nanoscale electronic, optoelectronic, electrochemical, and electromechanical devices. 1-D nanostructures have been synthesized for a lot of materials, including metals, II–VI and III–V semiconductors, sulfides, nitrides, etc., using a variety of synthetic techniques, such as solution process, vapor–solid process, vapor–liquid–solid process, template-directed process and so on [6-40].
Semiconducting II–V compounds are important narrow band gap semiconductors with great scientific and technological importance . They are suggested to exhibit pronounced size quantization effects due to the large excitonic radii. Bulk II–V semiconductors have been used as infrared detectors, lasers, solar cells, ultrasonic multipliers, and Hall generators [42-50]. However, research on nanoscale II–V semiconductors, especially 1-D nanostructures, has been lingering far behind compared with the significant progress in the studies of 1-D II–VI and III–V semiconductors, mainly due to the significant synthetic experimental difficulties, such as lack of generalized synthetic methodologies, instability in air, etc. Since the first successfully synthesized trumpet-like Zn3P2 nanowires in 2006, many kinds of interesting 1-D II–V semiconductors nanostructures have been reported using different techniques, which greatly promote their further application in nanoscale electronic and optoelectronic devices.
This article will provide a comprehensive review of the state-of-the-art research activities focused on synthesis and devices of 1-D II–V semiconducting nanostructures. The first section introduces typical 1-D nanostructures obtained on II–V semiconductors, including nanotubes, nanowires, nanobelts, and some special nanostructures. Next, some important electronic and optoelectronic devices built on 1-D II–V semiconducting nanostructures are presented, which include metal–insulator-semiconductor field-effect transistors (MIS-FET), metal-semiconductor field-effect transistors (MS-FET), andp–nheterojunction photodiode. This review will then conclude with some personal perspectives and outlook on the future developments in the 1-D II–V semiconducting nanostructures research area.
Since the first successfully synthesized trumpet-like Zn3P2nanowires in 2006, many kinds of interesting 1-D II–V semiconducting nanostructures, such as nanotubes, nanowires, and nanobelts, special nanostructures have been reported using different techniques. In this section, we will discuss several typical 1-D nanostructures obtained on II–V semiconductors.
Considerable attention has been paid on semiconducting nanotubes since the first report of carbon nanotubes by Iijima in 1991 . Nanotubes are often obtained for materials with layered or pseudolayered structures . Without these kinds of structures, templates (hard templates or soft templates) are usually used to promote the formation of tubular structures [21,51]. Recently, we developed an in situ nanorod template method to synthesize high-quality single crystalline II–V nanotubes, including Zn3P2 nanotubes and Cd3P2 nanotubes . The whole process can be expressed as shown in Fig. Fig.1.1. This process was performed in a vertical induction furnace at high temperature and the mixture of ZnS (or CdS), Mn3P2 and P powders were used as the evaporation sources. At high reaction temperature, ZnS or CdS was evaporated and reacted with graphite to generate Zn or Cd vapors, which were transferred to low temperature region and deposited on the surface of graphite crucible, resulted in the formation of Zn or Cd nanorods due to the anisotropic nature of wurtzite Zn or Cd phases. At the same time, Mn3P2 was thermally decomposed to generate P gases. These newly generated P gases deposited on the surface of Zn or Cd nanorods and reacted with them to generate the Zn3P2 or Cd3P2 shells. After a reaction time, the inner Zn or Cd nanorods were consumed and finally only Zn3P2 or Cd3P2 nanotubes were formed.
Figure Figure2a2a is a SEM image of the Cd3P2product obtained in the in situ nanorod template process. 1-D nanostructures can be found deposited on the graphite crucible on a large scale. A magnified SEM image shown inset clearly reveals that they are hollow nanotubes. Figure Figure2b2b and c show typical TEM images of the obtained Cd3P2nanotubes. The clear brightness contrast confirms that they are hollow nanotubes. Typical nanotubes have diameters of about 100 nm and the wall thickness of several nanometers. As-synthesized Cd3P2nanotubes are of single crystalline nature with the preferred growth direction perpendicular to the (101) planes as revealed in Fig. Fig.2d.2d. Similarly, the Zn3P2products obtained in the in situ template process are also single crystalline nanotubes as shown in Fig. Fig.2e.2e. We should mention that defects existed in the as-synthesized Cd3P2and Zn3P2nanotubes because of the release of built-in strain.
Nanowire is one of the most common 1-D structure and many kinds of materials can form into the nanowire structures. Till now, Cd3P2, Zn3P2, and Cd3As2 among the II–V group semiconductors were found to be able to form into the nanowire structures. Omari et al.  reported the synthesis of single-crystalline Cd3As2 nanowires by thermal evaporation of Cd3As2 powders at 750 °C. Figure Figure3a3a is a SEM image of the Cd3As2 nanowires. Studies found that these Cd3As2 nanowires are single crystals with the growth directions along the (112) crystal planes. Omari et al. also investigated the optical properties of these nanowires and found that they have IR active direct type absorption transitions at 0.11, 0.28, and 0.54 eV, which make it possible to use these nanowires as low cost optoelectronic devices and photodetectors operating in the long wavelength range.
Liu et al.  reported the synthesis of Zn3P2 nanowires by the reaction between Zn and InP powders at 850 °C. These Zn3P2 nanowires are also single crystals and have typical diameters of about 100 nm and lengths of tens of microns, as revealed in Fig. Fig.3b.3b. During this process, Au was used as catalysts to direct the nanowire growth and it is obviously a vapor–liquid–solid (VLS) process.
We were able to synthesize high quality Zn3P2 and Cd3P2 nanobelts on a large scale using a method similar to the synthesis of Zn3P2 and Cd3P2 nanotubes . Zn3P2 and Cd3P2 nanobelts were synthesized by using a mixture of ZnS (or CdS) and Mn3P2 powders as the evaporation source. The reaction was performed at 1350 °C for about 1 h. Figure Figure4a4a is a SEM image of the synthesized Zn3P2 nanobelts with rectangular cross-sections, which have diameters of about 100–200 nm and thickness of 40 nm. A TEM image of a single Cd3P2 nanobelt is depicted in Fig. Fig.4b.4b. The twisted structure confirms it is a nanobelt and the thickness is about 70 nm. The microstructures of the Cd3P2 nanobelts were studied by HRTEM and SAED. Figure Figure4c4c and d are the HRTEM images taken from the surface and body of a Cd3P2 nanobelt, respectively. The clearly resolved lattice fringes perpendicular to and along the longitudinal axis of the nanobelt are 0.71 and 0.87 nm, respectively, corresponding to the (101) and (010) lattice planes of tetragonal Cd3P2 phase. Studies revealed that these Cd3P2 nanobelts have the preferred growth directions along the  crystallographic orientations. For the synthesis of Zn3P2 and Cd3P2 nanobelts, no catalyst was used, indicating that it was governed by the vapor–solid (VS) mechanism. In fact, we found that Zn3P2 and Cd3P2 nanobelts can also be synthesized via the VLS mechanism if suitable catalysts, such as indium, were selected.
Synthesis and assembly of 1-D nanostructures with special morphologies, shapes, and compositions have attracted great interests very recently because they may process interesting physical and chemical properties associated with their specific characteristics. They may also be used to fabricate special electronic and optoelectronic devices which cannot be fulfilled using simple 1-D nanostructures.
Wang et al.  recently succeeded in synthesizing three-dimensional (3-D) branched tree-like Zn3P2 nanostructures as shown in Fig. Fig.5.5. These 3-D Zn3P2 nanostructures were synthesized in a thermal-assisted pulse-laser-deposition (PLD) system using Zn3P2/ZnO/Zn as the target. A top view image shown in Fig. Fig.5a5a demonstrated that these 3-D nanostructures have a sixfold symmetry. The tree-like branched shapes were confirmed by the side view SEM image depicted in Fig. Fig.5b.5b. The constituent branches of the tree-like structures were found to be as long as several tens of micrometers, with the diameters lying in the range of ten to several hundred nanometers. During this process, it was found that nanobelts were formed once the branches grew much longer as indicated in Fig. Fig.5c.5c. The composition of the tree-like Zn3P2 structures was checked using energy-dispersive spectrometry (EDS) and the results confirm they are pure Zn3P2 (Fig. (Fig.55d).
1-D bicrystalline nanostructures have received great attention because of their peculiar structures and structure-related properties. Several kinds of 1-D bicrystalline nanostructures have been obtained for ZnS, InP, etc. [57-62]. We synthesized bicrystalline Zn3P2 and Cd3P2 nanobelts via a VS process in the vertical induction furnace system . The source materials used are a mixture of Zn (or Cd), ZnS (or CdS), GaP, and Mn3P2 powders. Figure Figure6a6a is a SEM image of the Zn3P2 product, which clearly shows that long and straight nanobelts are obtained on a large scale. Each nanobelt has a uniform width of 100–200 nm and length in the range of tens of micrometers. A high-magnification SEM image shown in Fig. Fig.6b6b clearly reveals that these nanobelts are bicrystals with distinct grain boundary along the direction parallel to the growth direction. TEM images of several typical Zn3P2 bicrystal nanobelts were depicted in Fig. Fig.6c6c and d. Grain boundaries were clearly observed in the middle of the nanobelts, indicating the bicrystal nature. These Zn3P2 bicrystal nanobelts are composed of two single-crystal with the preferred growth axis along the  directions. HRTEM image taken from a single Cd3P2 bicrystal nanobelt was shown in Fig. Fig.6e.6e. The marked lattice fringes in each part of the Cd3P2 nanobelt are 0.71 nm, in accordance with the (101) plane of tetragonal Cd3P2. The correspondence Fast Fourier Transform (FFT) patterns taken from the two parts within a single bicrystal nanobelt were demonstrated in Fig. Fig.6f6f and g, which verified the formation of bicrystalline Cd3P2 nanobelts with the preferred growth along the  directions.
Besides the above discussed special 1-D II–V semiconducting nanostructures, we were also able to get other kinds of interesting 1-D II–V semiconducting nanostructures. For instance, Fig. Fig.7a7a and b show the SEM and TEM image of zigzag twinned Zn3P2 nanowires, which were also obtained in the vapor phase process . Spherical indium nanoparticles were found attached to the zigzag nanowires, indicating that the process was governed by the VLS mechanism. These zigzag Zn3P2 nanowires typically have periodic twins with a period of 50–120 nm along the whole nanowires. Lots of defects were found in the twin boundaries area, similar to previous reports on twinned nanowires [57-62]. If we did not use indium catalysts and kept other conditions constant, we synthesized zigzag single crystal Zn3P2 nanowires as shown in Fig. Fig.7c7c and d. The angle between two neighboring kinks is about 120°, consistent with that between the (001) and (101) planes of tetragonal Zn3P2 phase. Figure Figure7e–g7e–g are SEM images of trumpet-like Zn3P2 nanowires, which are composed of a hollow cone supported by a nanowire . The trumpet-like Zn3P2 nanowires are single crystals with preferred growth directions along the  orientations. Other special II–V semiconducting 1-D nanostructures include 1-D hierarchical Zn3P2/ZnS nanotube/nanowires heterostructures, which consist of main Zn3P2 nanotube wrapped with high density ZnS nanowires .
We have discussed above several kinds of 1-D II–V semiconducting nanostructures obtained till now. By carefully controlling the experimental parameters, such as evaporation sources, temperature, carrier gases, etc., more 1-D nanostructures are expected to be obtained for II–V group semiconductors.
As an important group of narrow band gap semiconductors, 1-D II–V semiconducting nanostructures can be used to fabricate nanoscale electronic and optoelectronic devices. Several kinds of nanodevices had been fabricated built on single 1-D II–V semiconducting nanostructure, such as MIS-FET, MS-FET, andp–nheterojunction photodiode.
Figure Figure8a8a inset is a schematic illustration of a MIS-FET built on a single Zn3P2nanowire. Basically, the MIS-FET is supported on an oxidizedp-type silicon substrate with the underlying conducting silicon as the back gate electrode to vary the electrostatic potential of the nanostructure. Two metal contacts, such as Au and Ti/Au, corresponding to the source and drain electrodes, are defined by either photolithography or electron beam lithography, followed by evaporation of suitable metal contacts. TheIds–Vdscurves of the MIS-FET are illustrated in Fig. Fig.8a,8a, showing typicalp-type behavior of the Zn3P2nanowire. From theIds–Vdsmeasurement, the resistivity of the nanowire is calculated to be about 1.96 Ωcm. We also built MIS-FET using zigzag Zn3P2nanowire as shown in Fig. Fig.8b8b inset. As-fabricated MIS-FET also confirms thep-type behavior of the zigzag nanowire.
To investigate the electronic transport behaviors of 1-D II–V semiconducting nanostructures, we fabricated MIS-FET built on single Zn3P2 and Cd3P2 bicrystal nanobelts and explored the electronic transport behaviors as a function of temperature in vacuum . A SEM image of the Zn3P2 MIS-FET is depicted in Fig. Fig.8c8c inset. Figure Figure8c8c displays the I–V curves of a Zn3P2 bicrystal nanobelt MIS-FET device measured in the temperature region of 100–300 K without applying gate voltage. The conductance of the device continuously decreased as the temperature decreased. The zero-bias conductance at 300 K is calculated to be 27.75 nano-Siemens (nS) and it decreases to 0.01 nS at 100 K. Plotted the zero-bias conductance in a logarithmic scale as a function of 1000/T gives a linear behavior within the temperature region investigated. All these results suggested that the thermal activation of carriers is the dominant transport mechanism for the Zn3P2 bicrystal nanobelt MIS-FET. The electronic transport behavior of Cd3P2 bicrystal nanobelts was also investigated at different temperatures and the results (Fig. (Fig.8d)8d) also suggested a dominant thermal activation of carriers transport mechanism.
Liu et al.  fabricated MS-FET using single Zn3P2 nanowire as the active material. Basic structure of the MS-FET is demonstrated in Fig. Fig.9a,9a, in which Ni/Au was used as electrodes and Al as the top gate electrode across the nanowire. The typical ISDVSD characteristics of a p-type Zn3P2 nanowire based MS-FET measured at room temperature under gate biases ranging from −0.5 to 0 V with a step of 0.1 V is depicted in Fig. Fig.9b.9b. As-fabricated MS-FET shows p-type conductance behavior and it is turned off at zero gate bias, indicating that the MS-FET was in the E-mode.
Very recently, Wang et al. fabricated pn heterojunction photodiode using a crossed heterojunction made of p-type Zn3P2 nanowire and n-type ZnO nanowire . The device structure is displayed in Fig. Fig.10a10a inset and the I–V curve at reverse and forward bias is illustrated in Fig. Fig.10.10. Wang et al. studied the device behaviors by checking the photodiode under reverse bias and in the dark and under the illumination of light with wavelength of 532 or 680 nm. The results are presented in Fig. Fig.10b,10b, which show apparent current enhancement by the light. The increase of the current results from the generation of electron-hole pairs inside the depletion region and nearby under the excitation of the light. The pn heterojunction photodiode gives rapid response and very high on/off ratio upon light illumination.
In conclusion, we provide a comprehensive review of the state-of-the-art research activities focused on the synthesis and device applications of 1-D II–V semiconducting nanostructures. The rapid expended achievements, till now, toward 1-D II–V semiconducting nanostructures should inspire more and more research efforts to address the remaining challenges in this interesting filed.
Although comprehensive efforts have been made toward the synthesis of high-quality 1-D II–V semiconducting nanostructures, there is still plenty of room left unexploited. We believe that future work should continue to focus on generating them in more controlled, predictable, and simple ways. The II–V semiconductors exhibit pronounced size quantization effects due to the large excitonic radii, thus, it is important to synthesize 1-D II–V semiconducting nanostructures with diameters smaller than the excitonic radii. For example, one needs to find ways to get II–V semiconducting nanotubes with either very small diameter or very thin wall thickness. The physical and chemical properties of II–V semiconducting nanostructures with diameters smaller than the excitonic radii will then need to be investigated and more interesting results are expected to be gotten soon.
Besides, more functional nanoscale electronic and optoelectronic devices are expected to be built on 1-D II–V semiconducting nanostructures and the performance of the devices will be largely improved with the progress of producing high-quality 1-D II–V semiconducting nanostructures.
The authors acknowledge financial support from the High-level Talent Recruitment Foundation of Huazhong University of Science and Technology. The authors acknowledge the permission from the corresponding publishers/groups to reproduce their materials, especially figures, used in this paper.