Chapter 3 Synthesis of nanowires
3.5 Analysis results
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
20
3.5 Analysis results
In order to analyze a single nanowire by transmission electron microscopy (TEM), the wire was suspended on a measurement platform so that electron beam can penetrate the wire. The wire is divided into three parts. The end close to the heater is defined as top part. The end away from the heater is defined as bottom part. Between the top and the bottom is the middle part. To see if the wire is well crystalized or not and the growth orientation of the nanowire, selected area diffraction (SAD) was taken.
To see the distribution of the bismuth, antimony and telluride in the wire, EDX line-scan profile was taken. To know the ratio between the three element EDX point scan has been done.
Nanowire No.1
Figure 3.19 SEM image of a suspend nanowire No.1 which grown from Bi0.5Sb1.5Te3 film after annealing at 500 ℃for 5 days. The nanowire is 150 nm in diameter. The electrodes had already deposited by the FIB.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
21
Figure 3.20 TEM image of the nanowire No. 1.
Figure 3.21 Selected area diffraction pattern of the nanowire No. 1.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
22
Figure 3.22 The scanning TEM image of (a) top (b) middle (c) bottom part of the nanowire No.1. The EDX line-scan profile show that Bismuth, antimony and telluride homogeneously distributed through the nanowire.
(a)
(b)
(c)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
23
Figure 3.23 EDX point-scan spectrum of the (a) top (b) middle (c) bottom part of the nanowire No.1. The inset shows the corresponding point.
Table 3.5 Weight percentage and atomic percentage of three parts of the nanowire
Element Atomic% Atom number
Top Middle Bottom Top Middle Bottom
Bi 13.13 13.20 12.87 0.62 0.62 0.62
Sb 29.53 28.88 28.79 1.38 1.38 1.38
Te 57.52 57.92 58.34 2.71 2.71 2.80
(a)
(b)
(c)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
24
Nanowire No.2
Figure 3.24 SEM image of a suspend nanowire No.2 which grown from Bi1.5Sb0.5Te3 film after annealing at 490 ℃for 5 days. The nanowire is 220 nm in diameter.
Figure 3.25 TEM image of the nanowire No. 2.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
25
Figure 3.26 TEM image of the nanowire No. 2.
Figure 3.27 Selected area diffraction pattern of the nanowire No. 2.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
26
Figure 3.28 The scanning TEM image of (a) top (b) middle (c) bottom part of the nanowire No.2. The EDX line-scan profile show that Bismuth, antimony and telluride homogeneously distributed through the nanowire.
(a)
(b)
(c)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
27
Figure 3.29 EDX point-scan spectrum of the (a) top (b) middle (c) bottom part of the nanowire No.2. The inset shows the corresponding point.
Table 3.6 Weight percentage and atomic percentage of three parts of the nanowire
Element Atomic% Atom number
Top Middle Bottom Top Middle Bottom
Bi 37.54 37.34 37.46 1.66 1.66 1.66
Sb 7.64 7.74 7.57 0.34 0.34 0.34
Te 54.83 54.92 54.97 2.43 2.44 2.44
(a)
(b)
(c)
‧
conductivity, nanowire was suspended on the measurement platform to allow the temperature fluctuation. In order to measure the thermoelectric properties, electrodes, heater and thermometers were fabricated on the measurement platform. Secstion4.1 introduces the acquired equipment and techniques. Section 4.2 shows how to fabricate the measurement platform and suspend a wire on it. Section 4.4 shows how to measure the thermoelectric properties of the nanowire. Section 4.5 shows the measurement result.4.1 Experimental equipment and techniques Photolithography
Photolithography is a process used in microfabrication to selectively remove parts of a thin film or the bulk of a substrate. It uses light to transfer a pattern from a mask to a light-sensitive chemical photoresist on the substrate. A series of chemical treatments then either engraves the exposure pattern into, or enables deposition of a new material in the desired pattern upon, the material underneath the photo resist.
Dry etch
Dry etching refers to the removal of material, typically a masked pattern of semiconductor material, by exposing the material to a bombardment of ions that dislodge portions of the material from the exposed surface. Unlike with many of the wet chemical etchants used in wet etching, the dry etching process typically etches directionally or anisotropically.
‧
good process control. Etching a (100) silicon surface through a rectangular hole in a masking material creates a pit with flat sloping <111>-oriented sidewalls and a flat<100>-oriented bottom. The <111>-oriented sidewalls have an angle to the surface of the wafer of: tan−1√2 = 54.7°. If the original rectangle was a perfect square, the pit when etched to completion displays a pyramidal shape.
Lift-off process [7]
A polymer resist layer is patterned first by optical or e-beam lithography.
Metallic thin film is then deposited onto the patterned resist layer. A wet chemical solution dissolves the resist layer, which also lifts off the metallic thin film on top of resist layer from the substrate. Only the metallic film deposited through the resist pattern opening onto the substrate remains. In this way, the resist pattern is transferred onto the substrate as a metallic pattern of reverse polarity.
Focused ion beam (FIB)
FIB systems operate in a similar fashion to a scanning electron microscope (SEM) except, rather than a beam of electrons and as the name implies, FIB systems use a finely focused beam of ions (usually gallium) that can be operated at low beam currents for imaging or high beam currents for site specific sputtering or milling. An FIB can used to deposit material via ion beam induced deposition. FIB-assisted chemical vapor deposition occurs when a gas, such as tungsten hexacarbonyl (W(CO)6) is introduced to the vacuum chamber and allowed to chemisorb onto the sample. By scanning an area with the beam, the precursor gas will be decomposed into volatile and non-volatile components; the non-volatile component, such as tungsten, remains on the surface as a deposition.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
30
Probe station with micropositioner
A probe station can be used to physically acquire signals from the internal nodes of a semiconductor device. The probe station utilizes manipulators which allow the precise positioning of thin needles on the surface of a semiconductor device. Here, the setup is used for manipulating nanowires. The micropositioner is equip with cat-whisker probe tip and fixed on probe station.
Figure 4.1 Set up of probe station with micropositioner for manipulating nanowire.
Four-point probe method
Current is supplied via a pair of current leads generate a voltage drop across the specimen and also across the current leads themselves. To avoid including that in the measurement, a pair of voltage leads is connected to the specimen. The accuracy of the technique comes from the fact that almost no current flows in the sense wires, so the voltage drop V=RI is extremely low.
Figure 4.2 Four-probe configuration for measuring the resistivity of a wire.
‧
Resistance thermometer is sensor used to measure temperature by correlating the resistance of the resistance thermometer element with temperature. The temperature dependence of electrical resistance of conductors is to a great degree linear and can be described by the approximation below:
ρ(T) = 𝜌0[𝛼0(𝑇 − 𝑇0)] 𝛼0 = 1 𝜌0[𝛿𝜌
𝛿𝑇]
𝑇=𝑇0
ρ0 just corresponds to the specific resistance temperature coefficient at a specified reference value. That of a semiconductor is however exponential:
ρ(T) = 𝑆𝛼𝐵𝑇
where S is defined as the cross sectional area and α and B are coefficients determining the shape of the function and the value of resistivity at a given temperature.
3ω method for thermal conductivity measurement [8]
In this method, either the specimen itself serves as a heater and at the same time a temperature sensor, if it is electrically conductive and with a temperature-dependent electric resistance. Feeding an ac electric current of the form 𝐼0sin 𝜔𝑡 into the specimen creates a temperature fluctuation on it at the frequency 2ω, and accordingly a resistance fluctuation at 2ω. This further leads to a voltage fluctuation at 3ω across the specimen.
Consider a uniform rod- or filament-like specimen in a four-probe configuration as for electrical resistance measurement. The two outside probes are used for feeding an electric current, and the two inside ones for measuring the voltage across the specimen. The specimen in between the two voltage probes is suspended to allow the temperature fluctuation. All the probes have to be highly thermal conductive, to heat sink the specimen at these points to the substrate. The specimen has to be maintained
‧
in a high vacuum and the whole setup is heat shielded to the substrate temperature to minimize the radial heat loss through gas convection and radiation.
Figure 4.3 Illustration of the four-probe configuration for measuring the specific heat and thermal conductivity of a wire.
In such a configuration and with an ac electrical current of the form 𝐼0sin 𝜔𝑡
where Cp, κ, R, and ρ are the specific heat, thermal conductivity, electric resistance and mass density of the specimen at the substrate temperature T0, respectively.
𝑅′= (𝑑𝑅/𝑑𝑇)𝑇0. L is the length of the specimen between voltage contacts, and S the cross section of the specimen. Let ∆(𝑥, 𝑡) denote the temperature variation from T0.
i.e. ∆(𝑥, 𝑡) = 𝑇(𝑥, 𝑡) − 𝑇0, Equation (3-1-1) and (3-1-2) become
𝜕
𝜕𝑡∆(𝑥, 𝑡) − 𝛼 𝜕2
𝜕𝑥2∆(𝑥, 𝑡) − 𝑐 sin2𝜔𝑡 ∙ ∆(𝑥, 𝑡) = 𝑏 sin2𝜔𝑡 (4.3) where 𝛼 = 𝜅/𝜌𝐶𝑝 is the thermal diffusivity and b = 𝐼02𝑅/𝜌𝐶𝑝𝐿𝑆, c = 𝐼02𝑅′/𝜌𝐶𝑝𝐿𝑆 The temperature distribution along the specimen would be:
‧
characteristic thermal time constant of the specimen for the axial thermal process. ∆0 is only κ dependent. The information of Cp is included in the fluctuation amplitude of the temperature around the dc accumulation.By solving the partial difference equation, the resistance fluctuation can be expressed as term at low frequencies, the 3ω component can be express as
𝑉3𝜔(𝑡) ≈ − 2𝐼03𝐿𝑅𝑅′
𝜋4𝜅𝑆√1 + (2𝜔𝛾)2sin(3𝜔𝑡 − 𝜙) (4.6) The root-mean-square (rms) values of voltage across the specimen contains a 3ω component
𝑉3𝜔 ≈ 4𝐼3𝐿𝑅𝑅′
𝜋4𝜅𝑆√1 + (2𝜔𝛾)2 (4.7) By fitting the experimental data to this formula, we can get the thermal conductivity κ and thermal time constant γ of the specimen. The specific heat can then be calculated as
𝐶𝑝 = 𝜋2𝛾𝜅 𝜌𝐿⁄ 2 (4.8)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
34
4.2 Primary measurement platform fabrication
First, silicon (Si) wafer with Si3N4 on the both sides was covered with photoresist by spin coating. Then photoresist was exposed to a rectangular pattern of ultraviolet light. After exposure, soluble photoresist would be developed by the developer. The wafer was then put into the RIE system. The Si3N4 without the protection of the photoresist would be etched by the reactive-ion. Next, the wafer is immersed in a bath of sodium hydroxide solution (NaOH). Si that expose to NaOH would be etch and then create cavities. Wafer with Si3N4 membranes would be complete after stripping.
Figure 4.4 Schematic representation of making Si3N4 membrane: (Step 1) Substrate spin coat with photoresist. (Step 2) Photoresist be exposed to a pattern of ultraviolet light. (Step 3) Soluble photoresist be developed by the developer. (Step 4) Remove Si3N4 by dry etch. (Step 5) Create cavities and leave a Si3N4 membrane by wet etch. (Step 6) Strip the photoresist.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
35
Lift-off process was used to make the contact pads of the measurement platform.
Si wafer with Si3N4 membrane was cover with photoresist by spin coating. Then photoresist was exposed to a pattern of contact pads of ultraviolet light. After exposure, soluble photoresist would be developed by the developer. Use the evaporator to deposit Ni/Au and then lift-off the photoresist by acetone. The primary measurement platform would be ready to be used after lift-off.
Figure 4.5 Schematic representation of depositing the contact pads: (Step 1) Substrate spin coat with photoresist. (Step 2) Photoresist be exposed to a pattern of ultraviolet light. (Step 3) Soluble photoresist be developed by the developer. (Step 4) Deposit Ni/Au. (Step 5) Lift-off the photoresist.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
36
4.2 Nanowires suspension and completion of measurement platform
Several methods were used to suspend the nanowire and complete the measurement platform.
Method one
First, the primary measurement platform was immersed in the DI water and put into the ultrasonic cleaner. Then the Si3N4 membrane was broken by the ultrasonic wave to open a window in the primary measurement platform. Next, the nanowire was picked up by a cat–whisker probe tip which manipulated by a micropositioner under the optical microscope. Then the nanowire was suspended on the on the primary measurement platform and deposit the six electrodes by the FIB.
Figure 4.6 Schematic representation of suspend the nanowire and deposit the electrodes by method one. (1)Prepare a primary measurement platform with membrane. (2)Break the membrane by ultrasonic wave. (3)Suspend the wire. (4)Deposit electrode by FIB.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
37
Figure 4.7 SEM image of a suspended nanowire.
Method two
Put the nanowire on the primary measurement platform. Part of the nanowire was laid on the Si3N4 membrane. The resistance thermometers, current leads and voltage leads would be made by the electron-beam lithography. Two kind of pattern were used in the measurement. Next, the membrane was etched by the ICP or broke by the tip.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
38
Figure 4.8 Schematic representation of suspend the nanowire and deposit the electrodes by method two. (a) and (b) follow the same procedure but with different pattern. (1)Prepare a primary measurement platform with membrane. (2)Put the wire on the primary measurement platform.
(3)Make the thermometer and electrodes by lift-off process. (4)Remove the membrane.
Figure 4.9 SEM top view image of the suspended nanowire.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
39
Figure 4.10 SEM tilt view image of the suspended nanowire.
Method three
First, the resistance thermometers, current leads were made by the electron-beam lithography on the primary measurement platform. The Si3N4 membrane can break by the ultrasonic wave, reacting ion, plasma or tungsten tip to open a window. Next, the nanowire was hanged across two resistance thermometers with two ends of the wire attach to the current lead. As two electrodes of the thermometer was also the voltage lead of 4-point probes method, the contacts of the nanowire and thermometer would be covered with a layer of platinum which are deposited by the FIB to make a better contact and also the contact of current leads.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
40
Figure 4.11 Schematic representation of suspend the nanowire and deposit the electrodes by method three. (1)Make the thermometers on the primary measurement platform by the lift-off process. (2)Break the membrane by ultrasonic wave. (3)Suspend the wire. (4)Deposit a layer of platinum to cover the contact.
Figure 4.12 SEM image of the suspended nanowire.
‧
4.4 Thermoelectric properties measurement of the nanowire 4.4.1 Resistivity measurement
Four-point probe method was applies to measure the resistivity. Feed an AC current via a pair of current leads into the specimen and measure the root mean square of the voltage difference via a pair of voltage leads. According to V = IR and ρ = RA/ℓ where V, I, R, ρ, A and ℓ are voltage difference, current, resistance, resistivity, cross-section area of the wire and length between a pair of voltage leads respectively, one can get the resistivity of the nanowire.
4.4.2 Seebeck measurement
To get the Seebeck coefficient, temperature gradient is generated by heater across the sample and thermoelectric voltage that is generated by the Seebeck effect is measure. To generate the temperature gradient, the heater is placed at one end of the sample and an AC current with frequency 1ω with magnitude equals to I sin 𝜔𝑡 is applied to the heater. Heater would produce heat because of the Joule heating.
Because heat that produced by the heater is proportional to the square of the current multiplied by the electrical resistance of the wire Q ∝ (𝐼2sin2𝜔𝑡)𝑅 where Q is the heat that produced by the heater and R is the electrical resistance of the sample and sin2𝛼 = (1 − cos 2𝛼 2⁄ ), so the heater would be heated at frequency 2ω. As the heater is heated at frequency 2ω, the temperature fluctuation on the sample would be also at frequency 2ω. As a length of metallic wire or part of the sample is used as the sensor of the thermometer, temperature coefficient of electrical resistance of them are needed to be known at first. As temperature is fluctuated at frequency 2ω, resistance of the sensor would change at frequency 2ω. By apply a DC current to the sensor and measure the change of the voltage difference between the two end of the sensor at frequency 2ω by using lock-in amplifier, it would able to know the resistance change
‧
of the sensor. Already knowing the temperature coefficient of electrical resistance of the sensor, how much degree different been created between two end of the sample would be known. By knowing the temperature difference and also measuring the thermoelectric voltage of two end of the sample, Seebeck coefficient can be calculated by the formula: S = − △ V △ 𝑇⁄ .
4.4.3 Thermal conductivity measurement
3ω method was applied for the thermal conductivity measurement. The measurement setup is much like the setup of resistivity measurement. The specimen between the two voltage probes should be suspended to allow the temperature fluctuation. Feed an AC current of the form 𝐼0sin 𝜔𝑡 via a pair of current leads into the specimen and lock the V3ω signal via a pair of voltage leads. Theoretical calculation 𝑉3𝜔 ≈ 4𝐼3𝐿𝑅𝑅′⁄𝜋4𝜅𝑆√1 + (2𝜔𝛾)2. By fitting the experimental data to this formula, one can get the thermal conductivity κ and thermal time constant γ of the specimen. Further detail will shoe in the
There are two ways to perform the measurement. In the first, the measurement platform is maintained at fixed temperatures, and then the frequency dependence of V3ω is measured. In this way, we can check the I3 and the 1 √1 + (2𝜔𝛾)⁄ 2 dependencies of V3ω as well as the relation tan 𝜙 = 2𝜔𝛾. In the second way of measurement, the temperature of the measurement platform is slowly increase or decrease, and the working frequency of the lock-in amplifier is changed between a few set values. The maximum working frequency is adjusted by keeping 2ωγ<4.
‧
Several inner electrode pattern designs are use in the measurement.
Pattern one
For resistivity and thermal conductivity measure, electrodes A and B are current leads. Electrodes C and D are connected to a locking amplifier. For Seebeck measurement, part of the nanowire between the contact of electrode C and E is the high temperature sensor and part of the wire between the contact of electrode D and F is the low temperature sensor. Current via electrode A and B feed into the sensors.
Electrodes C and D is a pair of voltage lead for measuring the voltage difference that is generate by the Seebeck effect.
Figure 4.13 Schematic representation of pattern one
Pattern two
For resistivity and thermal conductivity measure, electrodes A and B are current leads. Electrodes C and D are connected to a locking amplifier to lock the V1ω and V3ω signal. For Seebeck measurement, a length of gold wire vertical to the heater between the contact of electrode G and H is the high temperature sensor of the thermometer Th and a length of gold wire between the contact of electrode I and J is the low temperature sensor of the thermometer Tc. Current via electrode E and F feed into the thermometer Th and Tc. Electrodes C and D is a pair of voltage lead for measuring the voltage difference that is generate by the Seebeck effect.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
44
Figure 4.14 Schematic representation of pattern two
Pattern three
For resistivity and thermal conductivity measure, electrodes A and B are current leads. Electrodes C and D are connected to a locking amplifier to lock the V1ω and V3ω signal. For Seebeck measurement, current via electrode C and E feed into the thermometer Th and via electrode D and F feed into the thermometer Th. A length of gold wire parallel to the heater between the contact of electrode H and G is the high temperature sensor of the thermometer Th and a length of gold wire between the contact of electrode I and J is the low temperature sensor of the thermometer Tc. Electrodes C and D is a pair of voltage lead for measuring the voltage difference that is generate by the Seebeck effect..
Figure 4.15 Schematic representation of pattern three
‧
Bi0.62Sb1.38Te2.74 nanowire with diameter 150nm was excited by a constant alternating current about 0.1μA, where it is a sine wave 𝐼0sin 𝜔𝑡 profile with constant frequency f=9.731Hz. The experimental data of resistivity in temperature range 3.5 – 300 K of the nanowire was shown in Figure 4.16. The corresponding voltage signal with less than two degree shift was picked up by the lock-in amplifier (Figure4.17).0 50 100 150 200 250 300
Figure 4.16 The resistivity of the Bi0.6Sb1.4Te3 nanowire with diameter 150nm.
0 50 100 150 200 250 300
Figure 4.17 The temperature dependence of the phase angle of V.
‧
Figure 4.18 The temperature difference ΔT dependence of the thermoelectric voltage ΔV at T=200K
Figure 4.19 The Seebeck coefficient of the Bi0.6Sb1.4Te3 nanowire with diameter 150nm.
‧
We applied the 3ω method to measure the thermal conductivity of a suspended Bi0.6Sb1.4Te3 nanowire with diameter 150nm by using the approximation solution:
We applied the 3ω method to measure the thermal conductivity of a suspended Bi0.6Sb1.4Te3 nanowire with diameter 150nm by using the approximation solution: