^{1,a)}, Jessica Bolinsson

^{2}, Maria E. Messing

^{2}, Sebastian Lehmann

^{2}, Jonas Johansson

^{2}and Philippe Caroff

^{3}

### Abstract

Crystal structure and defects have been shown to have a strong impact on III-Vnanowire properties. Recently, it was demonstrated that the issue of random stacking and polytypism in semiconductornanowires can often be controlled using accessible growth parameters (such as temperature, diameter, and V/III ratio). In addition, it has been shown that crystal phase can be tuned selectively between cubic zinc blende and hexagonal wurtzite *within* individual nanowires of III-V materials such as InAs. In order for such results to be generally applied to different growth setups, it is necessary to fully explore and understand the trends governing crystal phase dependencies on all accessible growth parameters, including how they relate to each other. In this study, the authors have systematically investigated the influence of temperature, diameter, V/III ratio, and total mass flow on the crystal structure of InAsnanowiresgrown by metal-organic vapor phase epitaxy over a broad parameter range. The authors observed that each of these accessible parameters can affect the resulting crystal structure, and that the trends for each parameter are affected by the magnitude of the others. The authors also noted that most of the parameter dependencies are nonlinear and, in fact, exhibit threshold values at which structure changes discontinuously. By optimizing each of the growth parameters, it is shown that pure ZB or pure WZ phase can be achieved for several different sets of growth conditions. The roles of nucleation kinetics, thermodynamics, and precursor chemistry are also discussed to compare the results to current nanowiregrowth models. The results in this work should facilitate comparison of data and transfer of knowledge between different growth systems and techniques, which, in turn, should lead to greater understanding of polytypism in nanowires and greater control and freedom in nanowire crystal phase engineering.

This work was supported by the Nanometer Structure Consortium at Lund University (nmC@LU), the Swedish Foundation for Strategic Research (SSF), the Swedish Research Council (VR), and the Knut and Alice Wallenberg Foundation.

I. INTRODUCTION

II. EXPERIMENT

III. RESULTS AND DISCUSSION

A. Qualitative description

B. Temperature and diameter effects

C. V/III ratio effects

D. Total mass flow effects

E. Relationship to current models

F. Role of precursor chemistry

IV. SUMMARY AND CONCLUSIONS

### Key Topics

- Nanowires
- 94.0
- Crystal structure
- 63.0
- III-V semiconductors
- 28.0
- Crystal defects
- 21.0
- Semiconductor growth
- 17.0

## Figures

TEM images showing crystal structure variations in InAs nanowires. (a) Predominantly ZB nanowire with WZ inclusions of several nanometer long. The c-plane proportion for this nanowire is 0.8. (b) Predominantly ZB nanowire with a high density of individual twin planes. The c-plane proportion for this nanowire is 0.8. (c) Nanowire containing a mixture of ZB and WZ phases with randomly distributed stacking faults and twin planes. The c-plane proportion is close to 0.5. (d) Pure ZB nanowire with c-plane proportion of 1, with the exception of a single WZ segment near the top interface (“neck region”). (e) Predominantly ZB nanowire containing twin planes, arranged in pairs such that there is a favored twin segment orientation. The c-plane proportion is 0.97.

TEM images showing crystal structure variations in InAs nanowires. (a) Predominantly ZB nanowire with WZ inclusions of several nanometer long. The c-plane proportion for this nanowire is 0.8. (b) Predominantly ZB nanowire with a high density of individual twin planes. The c-plane proportion for this nanowire is 0.8. (c) Nanowire containing a mixture of ZB and WZ phases with randomly distributed stacking faults and twin planes. The c-plane proportion is close to 0.5. (d) Pure ZB nanowire with c-plane proportion of 1, with the exception of a single WZ segment near the top interface (“neck region”). (e) Predominantly ZB nanowire containing twin planes, arranged in pairs such that there is a favored twin segment orientation. The c-plane proportion is 0.97.

(Color online) Proportion of ZB phase (counted as proportion of c-planes) vs nanowire diameter for different growth temperatures, at a V/III ratio of 100 and 130 , and TMI molar fraction of . Lines are guides for the eye. For all temperatures, there appears to be a transition from WZ to ZB with increasing diameter, such that there is a characteristic transition diameter for each temperature. In the temperature range of , the transition diameter decreases with temperature, but for temperatures below , the transition is shifted to a much lower temperature (such that no transition is visible for within the investigated diameter range).

(Color online) Proportion of ZB phase (counted as proportion of c-planes) vs nanowire diameter for different growth temperatures, at a V/III ratio of 100 and 130 , and TMI molar fraction of . Lines are guides for the eye. For all temperatures, there appears to be a transition from WZ to ZB with increasing diameter, such that there is a characteristic transition diameter for each temperature. In the temperature range of , the transition diameter decreases with temperature, but for temperatures below , the transition is shifted to a much lower temperature (such that no transition is visible for within the investigated diameter range).

(Color online) Proportion of ZB phase vs nanowire diameter for growth temperature of and TMI molar fraction of , at V/III ratios of 20 and 100 (90–110 plotted together). There is a clear difference in the dependence, with ZB proportion increasing with diameter at high V/III, but decreasing with diameter at low V/III.

(Color online) Proportion of ZB phase vs nanowire diameter for growth temperature of and TMI molar fraction of , at V/III ratios of 20 and 100 (90–110 plotted together). There is a clear difference in the dependence, with ZB proportion increasing with diameter at high V/III, but decreasing with diameter at low V/III.

(Color online) Proportion of ZB phase vs growth temperature for nanowires grown with TMI molar fraction of and V/III ratios of 30 and 100. All nanowires have a diameter of . There is a sharp transition from predominantly ZB to predominantly WZ phase at a temperature around . Above this temperature, the proportion of ZB decreases slowly. The transition is sharper for higher V/III ratio.

(Color online) Proportion of ZB phase vs growth temperature for nanowires grown with TMI molar fraction of and V/III ratios of 30 and 100. All nanowires have a diameter of . There is a sharp transition from predominantly ZB to predominantly WZ phase at a temperature around . Above this temperature, the proportion of ZB decreases slowly. The transition is sharper for higher V/III ratio.

TEM images illustrating the effect of V/III ratio on InAs nanowire crystal structure at different temperatures. (a) Predominantly WZ nanowires grown at , TMI of , V/III of 30, diameter of 38 nm, and ZB proportion of 0.28. (b) Pure ZB nanowires grown at , TMI of , V/III of 100, diameter of 45 nm, and ZB proportion of 1. (c) ZB nanowires with frequent twin planes grown at , TMI of , V/III of 20, diameter of 35 nm, and ZB proportion of 0.68. (d) Pure WZ nanowires grown at , TMI of , V/III of 100, diameter of 33 nm, and ZB proportion of 0. (e) Pure WZ nanowires grown at , TMI of , V/III of 23, diameter of 41 nm, and ZB proportion of 0. (e) WZ nanowires with frequent stacking faults grown at , TMI of , V/III of 46, diameter of 39 nm, and ZB proportion of 0.12.

TEM images illustrating the effect of V/III ratio on InAs nanowire crystal structure at different temperatures. (a) Predominantly WZ nanowires grown at , TMI of , V/III of 30, diameter of 38 nm, and ZB proportion of 0.28. (b) Pure ZB nanowires grown at , TMI of , V/III of 100, diameter of 45 nm, and ZB proportion of 1. (c) ZB nanowires with frequent twin planes grown at , TMI of , V/III of 20, diameter of 35 nm, and ZB proportion of 0.68. (d) Pure WZ nanowires grown at , TMI of , V/III of 100, diameter of 33 nm, and ZB proportion of 0. (e) Pure WZ nanowires grown at , TMI of , V/III of 23, diameter of 41 nm, and ZB proportion of 0. (e) WZ nanowires with frequent stacking faults grown at , TMI of , V/III of 46, diameter of 39 nm, and ZB proportion of 0.12.

TEM images illustrating the effect of precursor mass flow on InAs nanowire crystal structure for selected V/III ratios; TMI flows indicated on the images are . Nanowires are grown at . (a) Pure ZB nanowire grown at V/III of 120, TMI of , and diameter of 58 nm. (b) ZB nanowire with WZ segment inclusions of several nm in length, grown at V/III of 120, TMI of , and diameter of 61 nm. (c) ZB nanowire with frequent twin planes grown at V/III of 30 and TMI of . (d) WZ nanowire with occasional stacking faults grown at V/III of 30 and TMI of .

TEM images illustrating the effect of precursor mass flow on InAs nanowire crystal structure for selected V/III ratios; TMI flows indicated on the images are . Nanowires are grown at . (a) Pure ZB nanowire grown at V/III of 120, TMI of , and diameter of 58 nm. (b) ZB nanowire with WZ segment inclusions of several nm in length, grown at V/III of 120, TMI of , and diameter of 61 nm. (c) ZB nanowire with frequent twin planes grown at V/III of 30 and TMI of . (d) WZ nanowire with occasional stacking faults grown at V/III of 30 and TMI of .

Images illustrating the effect of precursor mass flow on InAs nanowire growth at a growth temperature of and V/III ratio of 44. The precursor flows are for and for multiplied by factors of (a) 1, (b) 2, (c) 3, and (d) 4. The insets of (a)–(d) are TEM images illustrating the crystal structure for each of the conditions. It is clear that the growth rate and lateral overgrowth scale with the precursor mass flow, but any effect on the crystal structure is minimal.

Images illustrating the effect of precursor mass flow on InAs nanowire growth at a growth temperature of and V/III ratio of 44. The precursor flows are for and for multiplied by factors of (a) 1, (b) 2, (c) 3, and (d) 4. The insets of (a)–(d) are TEM images illustrating the crystal structure for each of the conditions. It is clear that the growth rate and lateral overgrowth scale with the precursor mass flow, but any effect on the crystal structure is minimal.

Article metrics loading...

Full text loading...

Commenting has been disabled for this content