In this article, we present a comprehensive analysis of the hardware and software solutions required to enable frequency scalable load-pull test benches operating in the (sub)mm-wave frequency bands. First, the constraints arising from the harmonic (nonlinear) operation of mm-wave extender modules are discussed and analyzed. Then, different hardware solutions for signal generation and control, together with the specific software algorithms required to realize a frequency scalable load-pull test bench, are presented. The measurement setup key performances are analyzed in different frequency bands up to 500 GHz, i.e., waveguide bands from WR10 up to WR2.2. Finally, the load-pull measurements on an HBT device at 75 GHz and a two-stage differential PA at 135 GHz are presented to show the capability of the proposed test bench to characterize and optimize mm-wave nonlinear components.

@en

In this contribution we present the developments and current performance of calibration substrates manufactured on 150 mm Quartz wafers (675 μm thick) based on a CMOS process technology. The passive structures required to realize on-wafer vector network analyzer calibration standards are benchmarked against commercially available (i.e., Alumina) substrates. First, an analysis of the process stability is presented for both reflective and resistive impedances across the entire wafer (i.e., 24 dies). Full-wave EM simulations are employed to realize accurate calibration artefact models aiming to achieve state-of-the-art calibration accuracy. The calibration quality in finally benchmarked on an independent line realized in the back-end-of-line of a Silicon based technology up to 67.5 GHz.@en

In this contribution we present a method for estimating linearity performance of devices operating in the higher millimeter-wave region, under modulated signals and over different loading conditions. The proposed method uses the power dependent vector gain extracted during continuous-wave large signal (load pull) measurements. The EVM prediction capability of the method is benchmarked with experimental load pull data with realistic modulated signals (QAM16) in the 5 GHz (RF) and in the 26 GHz (5G) bands on a 22nm CMOS FD-SOI device. The EVM estimated by the model correlates to the load pull measurements under complex modulated stimulus and properly predicts the best loading condition for linearity. Finally, the proposed method is used to estimate the EVM performance (QAM16) and the optimal loading condition for a 22nm CMOS-SOI device operating in the higher millimeter-wave region, at 165 GHz.@en

In this contribution, we present the performances of an IQ mixer-based RF-interferometer module, called the HΓ-VNA, designed to be used as an add-on to VNAs to improve the measurement sensitivity and accuracy of DUTs presenting extreme impedances (|Γ|>0.8). The calibration procedure used to obtain accuracy improvement is presented, and allows to set both reference and system impedance to any selected one. The procedure is benchmarked by comparing it to a conventional 50 Ω short-open-load calibration performed on a 50-Ohm VNA. For this purpose, a custom on-wafer calibration kit has been realized, using a fused silica substrate, featuring calibration loads with very high impedances (between 3.5 kΩ and 7 kΩ). Experimental results show accuracy improvements when applying the proposed calibration technique to the measurement of high impedance resistors, for frequencies below 4 GHz.

@en

As the number of wireless applications increases every year, overcrowding the RF/microwave spectrum, research community and industry are gradually starting to dedicate more attention to the less exploited (sub)millimeter wave spectrum, spanning from 30 GHz to 1 THz. While the high frequency and large available bandwidth of the latter promises very fast communication and the space for countless new applications, the development of new devices working at high frequency is hampered by a series of challenges affecting both technology development and implementation. One of the bottlenecks in new technology development is the availability of accurate and reliable measurement techniques, to support the design and the model validation of both passive and active devices working at (sub)millimeter wave frequencies. As a matter of fact, the test and measurement market dedicated to sub-THz applications has presented small developments in the last decades, with the core instrumentation and measurement techniques still based on the same principles dedicated to lower frequency applications. This thesis is dedicated to the development of calibration and measurement techniques for the characterization of (sub)millimeter wave devices, allowing to bridge the gap between the current available measurement instrumentation and the new needs in the sub-THz range. This is done by mainly addressing two aspects: the development of advanced techniques and artifacts for the characterization of electronic devices in their native environment (i.e., on-wafer), and the implementation of measurement techniques allowing to characterize the small- and large-signal behavior of devices and circuits at (sub)millimeter wave, while overcoming the instrument-related challenges present at those frequencies. @en

This paper analyzes and accurately models the complex noise behavior of vector network analyzers (VNAs) when measuring large-mismatch devices and subsequently shows how the VNA measurement noise performance is enhanced through implementation of a high-speed, broadband, active RF interferometer module. The presented VNA noise model provides a solid framework, benchmarked by measurement data, to analyze existing RF interferometer approaches. The performance improvement of the proposed interferometer implementation is then benchmarked in terms of magnitude and phase stability of the renormalized impedance level. A test bench employing the novel add-on RF interferometer module is presented and demonstrated to achieve high-speed cancellation of the scattered wave over a broad frequency band. The first experiment shows ultralow noise in a 1-18 GHz broadband measurement of co-planar waveguide 0.5-Ω and 5-k Ω impedance standards. Employing the proposed hardware setup improves the noise uncertainty for the 5-k Ω impedance standard by a factor of 8 and 20 at 1 and 18 GHz, respectively. In the second experiment, a factor of 2 height-resolution enhancement is achieved in a scanning microwave microscope when the RF interferometer module is added to the instrument.

@en

In this paper we present the measurement procedure to achieve direct on-wafer absolute power calibration in VNA-based mm-wave setups. The proposed approach employs 28 nm CMOS n-channel MOSFET as the power calibration transfer device, providing sufficient responsivity up to 325 GHz. The square law conversion from mm-wave (power) to DC (voltage) through the CMOS device is employed to achieve a direct on-wafer power calibration. The use of the calibration transfer device allows for a (power) calibration procedure of a mm-wave measurement setup with zero extender movements, thus minimizing errors originating from cable movements, and reducing calibration time when compared to the standard, calorimeter based, procedure. The approach is experimentally benchmarked against the instrumentation power meters procedure in the WR5 band (140220 GHz), showing a maximum error propagated through the calibration equations, over the entire band and multiple devices, lower than 1 dB.

@en

In this contribution we analyze the impact of radiation losses due to multimode propagations in (single medium) calibration substrates. The impact of the complex modelling of the loss mechanism due to radiation mode is applied to the specific case of TRL on-wafer calibrations for mm-wave operation. A quantitative analysis based on 3D EM simulation is performed to provide guidelines on the material to be used as the calibration substrate, the backside conditions, and the accuracy that can then be expected. Finally experimental data providing qualitative indication of the quality of calibrations on different media are presented for the WR10 band.

@en

We present a performance analysis of passive THz components based on Microstrip transmission lines with a 2-μmthin plasma-enhanced chemical vapor deposition grown silicon nitride (PECVD SiNX) dielectric layer. A set of thru-reflect-line calibration structures is used for basic transmission line characterizations. We obtain losses of 9 dB/mm at 300 GHz. Branchline hybrid couplers are realized that exhibit 2.5-dB insertion loss, 1-dB amplitude imbalance, and -26-dB isolation, in agreement with simulations. We use the measured center frequency to determine the dielectric constant of the PECVD SiNx, which yields 5.9. We estimate the wafer-to-wafer variations to be of the order of 1%. Directional couplers are presented which exhibit -12-dB transmission to the coupled port and -26 dB to the isolated port. For transmission lines with 5-μm-thin silicon nitride (SiNx), we observe losses below 4 dB/mm. The thin SiNx dielectric membrane makes the THz components compatible with scanning probe microscopy cantilevers allowing the application of this technology in on-chip circuits of a THz near-field microscope.@en

In this paper, we present a thru-reflect-line (TRL) calibration/de-embedding kit integrated in the back-end-of-line of a SiGe technology, which allows direct calibration at the first metallization layer, thus moving the reference planes as close as possible to the intrinsic device. The proposed calibration/ de-embedding kit features capacitively loaded inverted CPW lines, allowing to reduce the losses arising from the conducting (i.e., silicon) substrate, by confining the propagating field in the low-loss dielectric layers. The structures have been designed specifically for (sub) mm-wave measurements, as complementary to conventional de-embedding kits used at lower frequencies. A simplified analytical model is presented to support the design of the capacitively loaded CPW lines. Results of a calibration/de-embedding kit, realized using a 130-nm BiCMOS technology are experimentally obtained in the WR-03 waveguide band. Worst case error bounds providing error below 5% in the entire WR3 band are demonstrated. Comparison between direct calibration and de-embedding using calibration transfer from fused silica are provided to highlight the improvements of the proposed approach in the mm-wave bands. Finally, the TRL calibration kit is employed for S-parameter measurements of a heterojunction bipolar transistor and the results are compared to the HiCUM model of the same transistor for verification@en

In this paper, we propose a method based on 3-D electromagnetic simulations, for the characteristic impedance extraction of transmission lines employed in TRL calibration, focusing on lines integrated in silicon technologies. The accuracy achieved with TRL calibrations using the proposed characteristic impedance extraction is benchmarked versus conventional approaches, with an emphasis on aluminum pads structures operating in the (sub) millimeter-wave range. The proposed method proves to be insensitive to common sources of error (i.e., large pad capacitance and inductive pad-to-line transitions), which affect the accuracy of characteristic impedance extraction based on measurements, especially as the testing frequency increases. First, direct on-wafer TRL calibrations are performed on uniform CPWs (i.e., with no pads discontinuities) to demonstrate how the proposed method performs as good as the calibration comparison method and outperforms calibration transfer approaches. Finally, the method is applied to a nonuniform CPW-based calibration kit, demonstrating how the proposed method provides accurate results, improving the calibration quality that can be achieved using the calibration comparison method when inductive pad-to-line transitions are present.

@en

In this paper we present a synthetic waveguide integrated in a commercial BiCMOS back-end-of-line, employing artificial dielectrics (ADs) to reduce the component size. The AD is realized by employing floating pillars using the various layers available in the technology, thus fulfilling metal density rule and boosting the effective permittivity of the host medium (i.e., SiO2) to 12.5. The impact of the AD translates in a 56% width reduction of the waveguide, for the same cutoff frequency. The structure provides 1.6dB of losses per guide wavelength (λg) at 300GHz and more than 30dB suppression below 180GHz, avoiding the need of filters in multiplication stages.@en

A fabrication technology to realize THz microstrip lines and passive circuit components is developed and tested making use of a plasma-enhanced chemical vapor deposition grown silicon nitride (PECVD SiNx) dielectric membrane. We use 2 μm thick SiNx and 300 nm thick gold layers on sapphire substrates. We fabricate a set of structures for thru-reflect-line (TRL) calibration, with the reflection standard implemented as a short through the via. We find losses of 9.5 dB/mm at 300 GHz for a 50 Ohm line. For a branchline coupler we measure 2.5 dB insertion loss, 1 dB amplitude imbalance and 21 dB isolation. Good control over the THz lines parameters is proven by similar performance of a set of 5 structures. The directional couplers show -14 dB transmission to the coupled port, -24 dB to the isolated port and -25 dB in reflection. The SiNx membrane, used as a dielectric, is compatible with atomic force microscopy (AFM) cantilevers allowing the application of this technology to the development of a THz near-field microscope.@en

In this contribution we analyze the definition of reference planes in probe-level calibrations. The removal of the probe type from the calibration definition is presented first analyzing the transition discontinuities and defining to which extent they have to be incorporated in the calibration error terms. Subsequently a commercial calibration, which is defined specific to a probe topology, is considered and its frequency dependent standard response are computed accurately via a 3D EM simulator. These probe independent standard definitions are then used to compare the accuracy achieved on the same structures, (i.e., CPW lines of different lengths) from two different probe topologies. Finally, the data from both probes are compared, using a worst bound metric, to the data achieved when using the probe specific calibration data showing an accuracy improvement for the probe independent approach, validating the improved identification of the reference plane proposed here.@en