Skip to main content

Table 1 Comparison of phase shifter platforms currently available in 220 nm SOI at 1550 nm. Eπ is the energy required to switch and hold the state. ∆T is the difference in temperature, and t[s] is the time of operation

From: Ultra-compact nonvolatile phase shifter based on electrically reprogrammable transparent phase change materials

 

\({\varvec{\varDelta }\varvec{n}}_{\varvec{e}\varvec{f}\varvec{f}}\)

IL(dB)

Lπ (µm)

EÏ€

VÏ€ (V)

Non-volatility

Rise timed

Thermo-optical

(doped-Si) [12]

~ 1.8 × 10− 4 Ă—âˆ†T[K]

0.23

61.6

24.8 Ă— t[s] mJ

4.36

No

2.69 µs

Thermo-optical

(metal heater) [12, 48]

~ 1.8 × 10− 4 Ă—âˆ†T[K]

< 0.4

> 200

20–25 × t[s] mJ

4

No

3.7 µs

Thermo-optical

(ITO) [49]

~ 1.8 × 10− 4 Ă—âˆ†T[K]

< 0.01

50

10 Ă— t[s] mJ

9.1

No

5.2 µs

Thermo-optical with slow light [50]

~ 0.1

2

~ 10

2 Ă— t[s] mJ

4.5

No

100 ns

Electro-optical (depletion) [51]

1.5 × 10− 4

3.6

1500

~ 2 × t[s] pJ a

10.7

No

15.5 ps

Electro-optical (injection) [52]

~ 10− 3

2

400

1.7 Ă— t[s] mJ

2

No

24 ns

Optoelectro-

mechanical [13, 15]

~ 0.01

0.47

84

179 Ă— t[s] nJ

2

No

123 ns

Liquid crystal [9]

0.016

0.25

49

1 Ă— t[s] nJ

5

No

1 ms

Plasmonic nonlinear polymer [10]c

0.055

12

29

4 Ă— t[s] mJ

45

No

15 ps

BTO Pockels effect [11]

7 × 10− 4

1

1000

8.2 Ă— t[s] mJ

3.3

No

11.7 ps

BTO Ferroelectric domain switching [53]

~ 1 × 10− 4 / pulse

0.07

150

4.6–26.7 pJ /pulse e

-10

Yes

300 ns

LiNbO3 [54]

1.5 × 10− 4

2.5

5000

53 Ă— t[s] mJ

5.1

No

3.5 ps

Transition metal dichalcogenides [55]

1.5 × 10− 3

0.55

1000

0.64 Ă— t[s] nJa

8

No

1 ns

30 nm Sb 2 Se 3 cladding cry/am (this work)

0.07

0.36b

11

38.4 µJ /

176 nJ

6.2/ 21.0

Yes

0.1-1 ms /

800 ns

  1. a No current flows in this device, making the power dissipation extremely low. However, a constant supply of large voltage means a significant power consumption of the electronics. b Based on a 12 μm-long Sb2Se3 OPS – it can be further decreased if using lower doping concentrations. c These values are calculated from the supplementary material of the referenced paper. d 10-90% rise time, where we used the single-pole approximation: \(\tau =0.35/{f}_{3dB}\) for results with no rise time reported. e Several pulses are required to achieve full π phase shift, and a constant DC signal is required to read out