Waveguide-Integrated, Plasmonic Enhanced Graphene Photodetectors

Abstract

We present a micrometer scale, on-chip integrated, plasmonic enhanced graphene photodetector (GPD) for telecom wavelengths operating at zero dark current. The GPD is designed and optimized to directly generate a photovoltage and has an external responsivity∼12.2V/W with a 3dB bandwidth∼42GHz. We utilize Au split-gates with a∼100nm gap to electrostatically create a p-n-junction and simultaneously guide a surface plasmon polariton gap-mode. This increases light-graphene interaction and optical absorption and results in an increased electronic temperature and steeper temperature gradient across the GPD channel. This paves the way to compact, on-chip integrated, power-efficient graphene based photodetectors for receivers in tele and datacom modules.The ever-growing demand for global data traffic[1] is driving the development of next generation communication standards [2,3]. The increasing numbers of connected devices[4], the need for new functionalities, and the development of high-performance computing [5,6] require optical communication systems performing at higher speeds, with improved energy-efficiency, whilst maintaining scalability and cost-effective manufacturing. Si photonics[7-9] offers the prospect of dense (nanoscale) integration[10] relying on mature, low-cost (based on complementary metal-oxide-semiconductor (CMOS) fabrication processes) manufacturing [8,9], making it one of the key technologies for short-reach (<10km) optical interconnects[11] beyond currently employed lithium niobate[12] and indium phosphate[13]. A variety of functionalities have been developed and demonstrated in Si photonics for local optical interconnects[11]. Electro-optic modulators based on carrier-depletion (phase-modulation) in Si[14, 15] or the Franz-Keldysh effect[16] (amplitude-modulation) in strained Si-Ge[17, 18] encode information into optical signals at telecom wavelengths (λ =1.3-1.6µm). On the receiver side, Ge[19] or bonded III-V[20, 21] photodetectors (PD) are needed for optical-to-electrical signal conversion, since the telecom photon energies are not sufficient for direct (band-to-band) photodetection in Si[22].On-chip integrated Ge PDs [23][24][25][26][27] are standard components in Si photonics foundries [8,9,22]. Their external responsivities (in A/W), R I = I ph /P in , where I ph is the photocurrent and P in is the incident optical power, can exceed 1A/W [8,23] and their bandwidth can reach 60GHz [25][26][27]. Following the development of high temperature (> 600 • C) [19] heterogeneous integration of Ge-on-Si using epitaxial growth and cyclic thermal annealing [19,28,29], the concentration of defects and threading dislocations in Ge epilayers and at Si/Ge interfaces can be reduced [19], resulting in low (<10nA[9, 27]) dark current in waveguide integrated Ge p-i-n photodiodes [24,27]. However, Ge-on-Si integration is a complex process [19,22,29], as the lattice mismatch between Si and Ge [19], ion implantation [23,25], thermal budget (i.e. thermal energy transfer to the wafer) management [22], and the non-plan...

Ultrafast Plasmonic Graphene Photodetector Based on the Channel Photothermoelectric Effect

“…Since the electron-electron scattering in graphene is ultrafast, on the order of 10 fs, and electron-phonon scattering relatively slow, on the order of picoseconds [49][50][51][52][53][54], the photoinduced carriers are thermalized with an effective temperature Te before they dissipate heat to the lattice. Typically the cooling length of hot electrons in graphene exceeds hundreds of nanometers, and so, the electrons in graphene would not reach thermal equilibrium with the lattice before being collected [26,[49][50][51][52]. As mentioned earlier, in the proposed design the electric field reaches its maximum at the metal stripe edge ("hot" electrode) and quickly decays in the direction of second electrodemleading to an asymmetric temperature distribution that can be transduced into an electrical signal through the Seebeck effect.…”

“…To find an answer to the questions raised above, we first need to classify the main effects that can contribute to photocurrent (or photovoltage) generation in waveguide-integrated graphene devices [25][26][27][28]. One can distinguish three main effects -photo-voltaic (PV) [29][30][31][32][33][34], photo-bolometric (PB) [35][36][37][38][39] and photo-thermoelectric (PTE) [40][41][42][43] effects.…”

“…Recent studies have shown that the PTE effect dominates over PV photocurrent generation in graphene devices [44,45], therefore it is the PTE devices that we consider optimizing in this work. The operational principle of most PTE detectors relies on a temperature gradient in the material's p-n junction under an electromagnetic field excitation, resulting in a net thermoelectric voltage across the material due to the Seebeck effect [16,21,26,33,[40][41][42][43]46]. Thus, the generated photocurrent is proportional to the Seebeck coefficient for the two sides of the junction.…”

“…As previously demonstrated, the main utility of the PTE effect is to creating an asymmetry in the device. Such an asymmetry can be created by using two different contact metals [29] or by using two adjacent graphene regions of different doping [26]. The photovoltage in the present case is generated at the junction and is driven by the difference in graphene's Seebeck coefficient ΔS=S1-S2 on either side of the junction through:…”

“…The graphene photodetectors have the additional advantage of process compatibility with standard photonic integrated circuits [22][23][24]. However, up to date, most of the graphene photodetectors suffer from either low responsivity [14,17,19,21,22,40], large dark currents associated with biasing the graphene channel [20,21,39] or complex design [26,39]. Thus, finding a better photodetector configuration that can takes full advantages of graphene's properties is highly desired.…”