Irradiation of new optoelectronic components for HL-LHC data transmission links

Candidate optoelectronic components for use in future data-transmission links at the High-Luminosity Large Hadron Collider (HL-LHC) were irradiated with 20 MeV neutrons at the University Cyclotron in Louvain-La-Neuve, Belgium and 24 GeV protons at the CERN PS irradiation facility. The results from this test for multi-channel transmitters, Germanium photodiodes, and Silicon photonics modulators are presented here.


Introduction
The upgrade from LHC to High-Luminosity LHC (HL-LHC) will increase the luminosity of the LHC by a factor of 5-10 thus imposing even more stringent requirements on the optoelectronic components used in the front-ends of data-transmission links at the HL-LHC. Not only must they operate in harsher radiation environments, but they must also cope with the increase in data-rates. Components with lower power consumption, higher-speeds, and smaller sizes than those currently used in the optical links installed at the LHC are being investigated as possible candidates for HL-LHC data transmission links. Radiation tests were carried out over the past year on silicon photonics modulators and multichannel transmitters to investigate their tolerance to levels of radiation similar to those expected at the HL-LHC, this paper will present the results of these tests.

Optoelectronic components under test
This section will present the devices under test and the procedure used to carry out the irradiation tests on these devices.

Silicon photonics modulators
Silicon photonics is a new technology approach to modulating light to transfer data at high speeds and with low power over an optical fiber. A silicon photonics optical circuit that could modulate, process and detect light signals could be of interest to HEP applications because of its potentially small size, high speed, low power consumption, and integrability with CMOS electronics and design tools. This is a new technology who's radiation response, to the best of our knowledge, is not known and therefore as a first test of their suitability for HEP applications we looked at the effect of radiation on some Si-based optoelectronic components. published design of silicon optical modulators is the pn. The main difference is the use of various waveguide geometries, and junction orientation [2][3][4][5][6][7][8][9][10][11][12]. The work presented in this paper reports on the latest experimental results of Mach Zehnder modulators with phase shifters of lengths 0.95 mm and 4.7 mm based on carrier depletion in a pipin diode. The optical modulators operate at 40 Gbit/s, with extinction ratio (ER) of 3.2 dB and 6.6 dB respectively. Optical losses at the 40 Gbit/s operating points are as low as 4.5 dB and 6 dB for 0.95 mm and 4.7 mm long active regions, respectively. In particular, in the case of the longer phase shifter, such results demonstrate a very good tradeoff between loss and ER, which can be successfully compared with previous reported 40 Gbit/s modulators [9,10]. This reduction of the optical loss can be attributed to the use of a pipin diode as a phase shifter. Indeed, it has been demonstrated that this structure delivers a good compromise between high efficiency, speed, and low optical loss [11]. The results present in this paper are an improvement to the silicon optical modulator demonstrated in [12], by using a self-aligned fabrication technique in order to achieve an accurate localization of the active p-doped region in the middle of the waveguide and thus, fabrication reliability.

Device design and fabrication
The silicon modulator is based on carrier depletion of a pipin diode (shown in Fig. 1(a)) embedded in an asymmetric Mach-Zehnder interferometer (MZI). The diode is composed of a pin junction that allows a reduction of the optical loss due to the presence of a nonintentionally doped region in the waveguide. A p-doped slit embedded in the intrinsic region of the junction is depleted when a reverse bias is applied to the device, leading to an increase of the effective index variation [4]. The pipin diode is designed in a single-mode silicon rib waveguide of width 420 nm, height 390 nm, etching depth of 290 nm, generating a slab of 100 nm. Waveguide dimensions are chosen to maximize the optical TE-mode confinement in the depleted region. To reduce the access resistance between the active region and the electrodes, only 800 nm on each side of the rib are etched and p++-and n++-doped regions are used, as illustrated in Fig. 1(a) [12]. Phase shifters of lengths 0.95 mm and 4.7 mm were fabricated. The modulator is biased using coplanar traveling-wave electrodes in ground-signal-ground (GSG) configuration as shown in Fig. 1(b). The GSG configuration was chosen to be compatible with the available RF probes in the experimental set-up. The width of the signal electrode, as well as the gap between the Mach-Zehnder silicon optical modulator samples were provided by Université Paris Sud for the study of their radiation hardness. A schematic of the modulator layout and modulator structure are shown in figure 1 and figure 2 respectively.
Modulation is achieved via the carrier depletion in the active region of the p-i-p-i-n diode structure [1,2], which is achieved by applying a reverse bias to the junction. Phase modulation is obtained from the change in the effective refractive index of the device induced by the change in carrier densities. To obtain amplitude modulation the waveguides are laid out in a Mach-Zehnder configuration which converts the phase modulation in one branch into an amplitude modulation of light passing through the entire modulator.
Two chips containing 11 Mach-Zehnder silicon optical modulators each were irradiated using the 24GeV proton beam at the CERN PS irradiation facility (IRRAD1). The devices were irradiated over a period of 4 days (3-7 July 2012) with 24GeV protons at an average flux of 3.4x10 9 protons/cm 2 /s to obtain a total fluence of approximately 1x10 15 cm -2 . During the irradiation the chips were placed on double-sided kapton tape in the middle of a cut-out made in the 5x5 cm cardboard holders provided by the radiation facility.
Because no optical fibers were coupled to these devices this test will only report on the change -  in the DC characteristics of the devices after irradiation. A light-tight probe station was used to measure the forward and reverse IV characteristics of all the modulators on each chip before and 11 days after the end of the irradiation period.

Transmitters and receivers
The multi-channel transmitters and germanium photodiodes, alongside a range of reference devices including edge-emitting (EELs) and vertical-cavity (VCSELs) lasers and InGaAs photodiodes, were irradiated with 20 MeV neutrons at the University Cyclotron in Louvain-La-Neuve, Belgium. The devices were irradiated for 24 hrs and received a total fluence ¡= 7.0x10 15 n/cm 2 . This corresponds to a neutron flux which is 3-4 orders of magnitude higher than that expected for tracking detector electronics at the HL-LHC where the total fluence over 10 years of operation is approximately 6x10 15 n/cm 2 . The static performance of the devices was monitored during the test and their recovery was monitored for 360 hrs after the end of the irradiation period using the set-up shown in figure 3a.
At the start of a measurement cycle the current generator shown in figure 3a was used to increase the bias current (I) applied to the active channel of devices under test (DUTs) from zero to a maximum value in constant increments; the light output (L) and the forward voltage (V) across the DUT were measured at every bias step. The L-I-V curve of the DUT was used to extract the slope efficiency (Eff) and threshold current (I th ) of the device.
Two types of multi-channel transmitters were irradiated; a 12 channel InGaAs VCSEL array and a Coarse Wavelength Division Multiplexing VCSEL. The InGaAs laser is a 12-channel array transmitting at 1060 nm with an optical package consisting of a ceramic substrate, a VCSEL array, a micro-lens array and a hardcover with a thin MT-female fiber fixed to the module by a clip. To the best of our knowledge this is the first test carried out on an InGaAs laser in high radiation environments. The other multi-channel transmitter tested was an optical sub-assembly that is capable of transmitting 4 data channels over one multi-mode fiber using Coarse Wavelength Division Multiplexing (CWDM). Such devices are of interest to HL-LHC applications because they could allow an increase in the fiber-bandwidth of the installed link system with no modifications to the currently installed fiber plants.
The germanium photodiodes were also irradiated in the same test, alongside reference InGaAs and GaAs photodiodes, using the set-up shown in figure 3b. Only IV measurements of the germanium devices were possible, there only the change in the dark current of these devices could be monitored during irradiation.   3 Results from radiation tests

Silicon photonics modulators
The irradiated modulator test chip contained eleven different modulator designs each with slightly different characteristics. The overall effect of the protons on the modulators, shown here in figure 4 for two different modulator structures from the same test chip, was to increase the leakage current in the p-i-p-i-n junction and increase the breakdown voltage of the devices. Exactly how the shape of the IV curve changed after the irradiation varied from device to device, however results were consistent across devices with similar structures from different chips. Previously X-ray irradiations (to 500kGy) of modulator chips of the same design and fabrication batch were carried out at CERN and showed little change in the I-V characteristics of the devices after irradiation. It was expected, and confirmed by this test, that the particle irradiation, which primarily causes displacement rather than ionization in the target material, would lead to greater changes in the I-V characteristics. Because of the difficulty in extrapolating from the change in the I-V curves the effect of radiation on the modulation properties of these devices, a future test is planned where similar pigtailed devices will be irradiated.

Multichannel transmitters
The L-I-V curves of three of the twelve channels of the InGaAs VCSEL and all four of the CWDM array's VCSELs were recorded during the neutron irradiation. The typical evolution of the L-I-V curves of a single channel of the InGaAs array and the CWDM VCSEL during irradiation is shown in figure 5. The devices' L-I-V curves show that they behave as expected during the irradiation [4]; as the transmitters are irradiated their slope efficiency decreases and their threshold current increases.     The L-I-V data collected during the test is used to track the changes in threshold current and slope efficiency during the irradiation period (figure 6) and any annealing that might take place after the end of the test (figure 7). Both figures also include results from two reference transmitters, a 1310 nm EEL and an 850 nm VCSEL, irradiated at the same time as the multichannel transmitters and which have been included in previous radiation tests [4].
The increase in the threshold currents (figure 6a) of both the InGaAs array and the CWDM VCSEL is comparable to that observed in the reference short-wavelength VCSELs but their recovery (figure 6b) is faster. One possible explanation for this faster recovery is that the average temperature of the multi-channel devices was higher than that of the single transmitters which can speed up the annealing process. Although the change in threshold current of the devices induced by the particle irradiation is comparable to that observed in other short wavelength VCSELs, figure 7 shows that the observed decrease in the slope efficiency of the multichannel transmitters is almost twice that of the reference VCSELs. Two possible explanations can be put forward for the decrease in the slope efficiency of the multi-channel transmitters: either it was caused by darkening of the passive components in the packaging of the device, or by an increase in the temperature of -5 -     the device due to the increased current densities required to operate the devices after the radiationinduced increase in threshold current. In the case of the CWDM device the latter is suspected due to the strong correlation between the recovery of its threshold current (figure 6b) and slope efficiency (figure 7b) during the annealing period, while for the InGaAs VCSEL array further testing is required to identify the cause and therefore more samples of the InGaAs VCSELs will be irradiated in an upcoming test.

Germanium photodiodes
Because the irradiated Ge photodiodes were not packaged in such a way as to allow access to the light path of the devices we can only report on the changes in the DC characteristics of the devices. Figure 8 shows the dark current of the photodiodes during the irradiation period. The leakage current of both the InGaAs and Ge photodiodes increased as the devices were irradiated, although the shapes of the two curves are different. In addition, the leakage current of the Ge diodes was several order of magnitudes smaller than that of its InGaAs counterpart.

Conclusions
This paper presents the results from radiation tests conducted on silicon photonics modulators, multichannel transmitters, and germanium photodiodes to investigate their tolerance to levels of radiation similar to those expected at the HL-LHC.
Before and after IV measurements carried out on proton-irradiated MZI modulator samples obtained from Université Paris Sud showed an increase in the leakage current and the breakdown voltage; however no measurements were made on the impact of radiation on the modulating properties of the device. Future tests where pigtailed modulator samples will be irradiated are currently under preparation. In addition to the silicon photonics modulators multichannel transmitters and germanium photodiodes were irradiated to HL-LHC like fluence levels using the 20 MeV neutron irradiation facility at Louvain-la-Neuve. The irradiated multichannel transmitters, a 12-channel InGaAs VCSEL array and a CWDM VCSEL-based device, behaved in a manner similar to a shortwavelength VCSEL irradiated at the same test and were still operational at the end of the test. The radiation induced increase in the dark current of the irradiated Ge photodiodes was smaller than that of InGaAs photodiodes also tested.