Passivation Layer on the Performance of QDIPs
The combination of edge thinning structure and surface passivation by atomic
layer deposition on the the performance of QDIPs is investigated in this chapter. Lai
et al. [12] has discovered that surface passivation by atomic layer deposition of Al2O3
can also effectively reduce surface leakage current and elevate operation temperature.
Therefore, the combination of edge thinning structure and surface passivation is
adopted in this chapter to block the surface leakage current of QDIPs and achieve
higher operation temperature.
4.1 Sample Preparation
The sample investigated in this chapter is QDIP75. The schematic device
structure of device D, the standard device with surface passivation layer Al2O3, device
E, with edge thinning at top contact layer and surface passivation layer Al2O3, and
device F, with edge thinning at quantum dot layer and surface passivation layer Al2O3
are shown in Fig. 4.1 (a) and (b), respectively. The fabrication process of the device is
shown in Fig. 4.2 After the edge thinning structure formation as indicated in Fig. 2.5
(a)~(h), the samples were dipped into diluted HF (1%) solution for 1 minute and
followed by diluted NH4OH (10%) solution for 1minute. The HF treatment can
remove the native oxide and the NH4OH treatment can leave the surface-OH
terminated which may ensure better quality of Al2O3 layer. Then the samples were
immediately transferred to an ALD reactor for Al2O3 deposition.10nm of Al2O3
dielectric material deposited by Atomic Layer Deposition (ALD) at 300o C was used
to coat the mesa edges surrounding the QDIP. Then the RTA process was used to
eliminate the fixed oxide charges at the surface between QDIP and Al2O3 layer. The
annealing rate is 200 oC/min up to the temperature of 550 oC and last for 5 minutes.
Then the third photolithography step was used to define the region for ohmic contact.
After the third photolithography step, 20 minutes hard baking was used to formalize
the pattern. Then wet etching using of dilute HF (HF:H2O=1:50) to remove the spare
52
Al2O3 which was deposited on the ohmic contact region. The etching time is about 90
seconds. After wet etching process, the Au-Ge-Ni/Au alloy was deposited by thermal
evaporation and followed by thermal annealing at 420o C for 2 minutes to form ohmic
contacts. Finally, the device with edge thinning structure and surface passivation layer
Al2O3 are completed.
(a) (b) (c) Device D Device E Device F Fig. 4.1 The schematic structure of devices (a) D, the standard device with surface passivation by ALD Al
2O
3,(b) E, the device with edge thinning structure at top contact layer and surface passivation by ALD Al
2O
3, (c) F, the device with edge thinning structure at quantum dot layer and surface passivation by ALD Al
2O
3.
Surface Passivation Edge Thinining
Fig. 4.2 The fabrication processes of devices with edge thinning
structure and surface passivation layer, (a) edge thinning structure
formation (as Fig. 2.5 (a)-(h) show), (b) dip in HF / NH
4OH, (c) surface
passivaiton layer Al
2O
3by ALD, (d) RTA, (e) photoresist coating,
(f) developing, (g) wet etching Al
2O
3by diluted HF, (h) metals
evaporation, (i) lift off.
54
4.2 Results and Discussion
Figs. 4.3 to 4.5 show the I-V characteristics of devices D, E, and F, respectively.
The background limited performance (BLIP) of all devices can be reckoned as 70 K.
Fig. 4.6 (a) and (b) display the comparison of dark I-V characteristics of devices
A and D at 20 K and 90 K, respectively. The dark I-V characteristics of devices A and
D are similar. Fig. 4.7 (a) and (b) display the comparison of dark I-V characteristics of
devices B and E at 20 K and 90 K, respectively. At 20 K, the dark current of device E
is 1 order of magnitude lower than that of device B at bias voltage beyond 1.5V and
below -1.5V. At 90 K, the dark current of device E is 1 order of magnitude lower than
that of device B at bias voltage beyond 1.2V and below -1.0V. Fig. 4.8 (a) and (b)
display the comparison of dark I-V characteristics of devices C and F at 20 K and
90 K, respectively. At 20 K, the dark current of device F is 2 orders of magnitude
lower than that of device C at bias voltage beyond 1.0V and below -1.0V. At 90 K, the
dark current of device F is 2 orders of magnitude lower than that of device C at bias
voltage beyond 0.5V and below -0.5V.
Fig. 4.9 (a) and (b) display the comparison of dark I-V characteristics of devices
D, E and F at 20 K and 90 K, respectively. At 20 K, the dark current of device F and
device E are 2 orders and 1 order of magnitude lower than that of device D ,
respectively at bias voltage beyond 1.0V and below -1.0V. At 90 K, the dark current
of device F is 1 order of magnitude lower than that of device D at bias voltage beyond
0.5V and below -0.5V and the dark current characteristics of devices E and D are
similar.
According to the comparison of dark I-V characteristics, it can be inferred that
the dark current reduction effect of surface passivation layer Al2O3 combined with
edge thinning structure is better than that of surface passivation layer Al2O3 only.
The surface passivation layer Al2O3 combined with edge thinning structure at
quantum dot layer can reduce dark current by 2 orders of magnitude, which is the
optimum improvement in this experiment.
Fig. 4.10 (a), (b) and (c) display the comparison of photo I-V characteristics of
devices A and D, B and E, and C and F, respectively. The photo I-V characteristics of
devices A and D are similar. The photo current of device E is 1 order of magnitude
lower than that of device B at bias voltage beyond 1.5V and below -1.5V. The photo
current of device F is 2orders of magnitude lower than that of device C at bias voltage
beyond 1.0V and below -1.0V.
Despite the combination of edge thinning structure and surface passivation layer
Al2O3 can reduce dark current by 1~2 orders of magnitude, the photo current is also
reduced simultaneously.
56
-2 -1 0 1 2
10
-1410
-1210
-1010
-810
-610
-410
-210
0C u rren t (A )
Voltage (V)
Device D Photo 20 K 70 K 100 K 130 K
Fig. 4.3 The I-V characteristic of device D.
-2 -1 0 1 2 10
-1410
-1210
-1010
-810
-610
-410
-210
0C u rren t (A )
Voltage (V)
Device E Photo 20K 70K 100K 130K
Fig. 4.4 The I-V characteristic of device E.
58
-2 -1 0 1 2
10
-1510
-1310
-1110
-910
-710
-510
-310
-1C u rren t (A )
Voltage (V)
Device F Photo 20K 70K 100K 130K
Fig. 4.5 The I-V characteristic of device F.
(a)
(b)
Fig. 4.6 The comparison of dark I-V characteristics of devices A
and D at T= (a) 20 K and (b) 90 K.
60
(a)
(b)
Fig. 4.7 The comparison of dark I-V characteristics of devices B
and E at T= (a) 20 K and (b) 90 K.
(a)
(b)
Fig. 4.8 The comparison of dark I-V characteristics of devices C
and F at T= (a) 20 K and (b) 90 K.
62
-2 -1 0 1 2
Fig. 4.10 The comparison of photo I-V characteristics of devices (a) A
(a)
(b)
(c)
64
Figs. 4.11 to 4.13 (a) and (b) display the responsivities of devices D, E and F at
20 K and the highest operation temperature, respectively. The operation temperature
of devices D, E and F are 130 K, 105 K and 135 K, respectively, which are all close to
that of device A, 120 K. It can be inferred that the combination of edge thinning
structure and surface passivation layer Al2O3 doesn’t significantly improve the
operation temperature of QDIPs in this experiment.
On the other hand, at 20 K, Device F can work at broader bias voltage range
(-2.6V~2.6V) than Device A does (-0.8V~1.8V). It’s because the current at higher
bias voltage is reduced by the combination of edge thinning structure at QD layer and
surface passivation layer Al2O3.
Fig. 4.14 (a) and (b) display the detectivities of devices A and F at bias voltage =
0.6V and 1.6V, respectively. In both bias voltages, the detectivities of devices A and F
are close to each other. It can be inferred that the combination of edge thinning
structure and surface passivation layer Al2O3 doesn’t significantly improve the
detectivity of QDIPs in this experiment.
3 4 5 6 7 8
66
(the highest operation temperature) at different biases.
3 4 5 6 7 8
(the highest operation temperature) at different biases.
68