123129 research outputs found
Sort by
Hierarchically porous carbon supports enable efficient syngas production in electrified reactive capture
Full text linkDirect-air capture (DAC) of CO2 often uses alkali hydroxides (e.g. KOH) as sorbent, and relies on an energy-intensive thermal CaCO3/Ca(OH)2 step to release CO2 and regenerate the alkali hydroxide. Reactive capture instead uses alkali carbonate post-capture liquid as feedstock, seeking to convert the captured CO2 to value-added products while regenerating the capture liquid. Here we investigate the origins of low prior performance in electrochemical reactive capture systems, finding that the catalyst becomes starved of CO2 even at moderate current densities leading to a rapid decline in faradaic efficiency (FE). We then study how the catalyst support can be redesigned to tackle this problem, and construct hierarchical carbon supports featuring interconnected mesopores and micropores, our purpose to increase the interaction between in situ generated CO2, i-CO2 – the limiting reagent – and the catalyst. We find that the attachment chemistry of the catalyst to the support is critical: only when we disperse and tether the molecular catalyst do we prevent catalyst aggregation and deactivation under bias. We report as a result carbonate electrolysis at 200 mA cm−2 at 2.9 V with FE of 47 ± 1% for CO, this corresponding to an energy efficiency (EE) to 2 : 1 syngas of 50% at 200 mA cm−2 when H2 is added using a water electrolyzer. This represents a 1.5× improvement in EE at this current density relative to the most efficient prior carbonate electrolysis reports. The CO FE remains above 40% at current densities as high as 500 mA cm−2, and all systems studied herein achieve < 1% CO2 in the outlet stream. The cradle-to-gate carbon intensity is lowered to −1.49 tonCO2 per tonsyngas as a result of the increase in EE, and a CO2-free tailgas stream is provided that minimizes separation costs.This work received financial support from Saudi Aramco Technologies Company under the Agreement No. SATC-2022-016. We thank Drs. Tirzah Abbott and Krysten Lauren Villalon from Northwestern University's NUANCE Center for their assistance with cross-sectional SEM analysis. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2023-00253850). The XAFS analysis was conducted with the support of the grant NRF-2019M3D1A1079309. Theoretical calculations of this work were supported through computational resources and staff contributions provided for the Quest high-performance computing facility at Northwestern University which is jointly supported by the Office of the Provost, the Office of Research, and Northwestern University Information Technolog
Nucleophilic Covalent Ligands Enable Simultaneous Surface Reconstruction and Passivation of Colloidal InSb Quantum Dots for Stable Short‐Wave Infrared Photodetectors
Full text linkIndium antimonide (InSb) colloidal quantum dots (CQDs) are promising candidates for short-wave infrared (SWIR) photodetectors due to their large Bohr exciton radius and tunable bandgap in the 0.6–1.3 eV range. However, the formation of metal oxides on InSb surfaces during synthesis impedes charge transport, necessitating CQD resurfacing strategies for integration into photodetectors. Previous reports achieved progress in device efficiency by resurfacing these CQDs with acid-halide sequential treatments, but the device operating stability remains unsatisfactory. Herein, we report a solution-phase strategy for surface reconstruction and passivation of InSb CQDs using sulfur-based nucleophilic covalent ligands. We find that short-chain thiol molecules remove surface metal oxides through nucleophilic attack and enable robust passivation of In and Sb via strong covalent bonds, whereas metal sulfides are less effective at oxide removal and passivation. Consequently, the thiolate-passivated CQDs exhibit a tenfold decrease in trap state density compared to controls and remain structurally and optically stable for 5 months. We demonstrate InSb CQD SWIR photodetectors that realize a high external quantum efficiency (EQE) of 28% at 1450 nm, with the highest operating stability among reported CQD SWIR photodetectors, retaining 95% of performance following 300 h of biased and illuminated operation.The authors would like to thank Larissa Levina, Elenita Palmiano, Remi Wolowiec, and Damir Kopilovic for their technical assistance. This work was supported by the Natural Sciences and Engineering Research Council of Canada (RPGIN-2017–06477) and the Canada Research Chairs (CRC-2017-00318). A.G. and M. Y. acknowledge Mitacs-Accelerate (IT30845), and NSREC-Discovery (RGPIN-2015-04703) for funding. XPS, TEM, SEM, and ToF-SIMS measurements in this work were performed at the University of Toronto's Open Centre for the Characterization of Advanced Materials (OCCAM). GISAXS patterns were collected at the BXDS Beamline at the Canadian Light Source (CLS). The CLS is funded by NSERC, the Canadian Institutes of Health Research, the Canada Foundation for Innovation, the Government of Saskatchewan, Western Economic Diversification Canada, and the University of Saskatchewan
Leveraging Learned Models for Robust Decision Optimization and Offline Reinforcement Learning
Full text linkRecent advances in machine learning (ML) have driven the adoption of predictive models for decision-making in data-rich environments in areas like natural language processing, computer vision, and autonomous systems. These models are valuable for decision optimization, predicting uncertain parameters, and for offline reinforcement learning (RL), approximating environment dynamics to improve policy learning. However, integrating learned models into decision-making poses challenges: (1) misalignment between prediction accuracy and decision quality, (2) distribution shift causing compounding errors for out-of-distribution data, and (3) limited generalization when encountering novel scenarios during deployment.
(1) To address the first challenge, we focus on decision optimization tasks where learned models predict uncertain parameters. Traditional training emphasizes predictive accuracy, which may misalign with decision objectives. The Smart Predict-then-Optimize (SPO) framework links predictions to decision optimization, but approximations are typically used due to the difficulty of optimizing the SPO loss. We introduce a novel symbolic argmin operator, enabling the first globally optimal solution for linear SPO problems, revealing substantial gaps in existing approximations.
(2) In offline RL, model-based approaches simulate environment dynamics to improve policy learning and generalization but risk compounding errors in underrepresented areas. We enhance the model-based value expansion (MVE) method, which combines model-based and model-free value estimates, by developing its Bayesian version, which derives a posterior over value estimates to produce conservative estimates based on model uncertainty. This mitigates overestimation, improving policy stability by ensuring models are used only when reliable.
(3) Offline RL agents face epistemic uncertainty—lack of knowledge about the environment’s behavior—when encountering novel states during deployment. We propose a model-based planning approach that incorporates epistemic uncertainty via a belief distribution over possible environment dynamics, allowing the agent to reason about uncertainty by considering multiple plausible models. The agent continuously updates its belief based on new observations, adapting its actions and improving offline-learned policies in real time to better handle situations with limited data and high epistemic uncertainty.
These contributions advance the integration of learned models into decision-making frameworks, improving alignment with decision objectives, managing model errors, and enhancing adaptivity through planning in offline RL, leading to more robust and reliable decision-making.Ph.D
Promoting Social Justice and Inclusion: Reflecting on the Identities of Teacher Leaders to Drive Effective Teaching and Leadership Strategies
Full text linkThis research examines how social justice teacher leaders can support diverse students and colleagues in Ontario schools. It also explores how social identity and social location can influence the practices of social justice teacher leaders. Drawing on the conceptual framework that depicts the interconnectedness of teacher leaders’ perceptions, social identity, social location, critical consciousness, and praxis, this study employs a qualitative research design, wherein I conduct semi-structured interviews with ten experienced social justice teacher leaders. The majority of these leaders belonged to minoritized groups in Southern Ontario schools and hold informal leadership positions. Using the constant comparative method, I identified codes and themes grounded in the literature on social justice and teacher leadership. Results revealed how participants were committed to creating inclusion in their classrooms and schools. They described a variety of effective and inclusive teaching and social justice strategies such as empowering student voices, engaging in critical conversations to examine power and privilege, and guiding students in exploring and understanding their social locations and identities. Participants identified several strategies to develop themselves as leaders, raise critical consciousness of colleagues through collaborative activities; and advocate for students and colleagues at the school. Working conditions also influenced their leadership and social justice work. Their leadership practices were profoundly shaped by their individual identities and social locations, whether minoritized or privileged, to inform their social justice-centered practices. This study expands the scope of teacher leadership studies by highlighting the crucial social justice work that social justice teacher leaders do within their classrooms and schools.Ph.D
AAC PT600 pinto dry bean
Full text linkAAC PT600 is a high yielding pinto dry bean (Phaseolus vulgaris L.) cultivar with an upright, indeterminate bush with short vine (Type IIa) growth habit, early maturity and large seed size. AAC PT600 had significantly improved lodging resistance and partial field resistance (i.e., avoidance) to white mould compared to both Island and AAC Expedition, pinto bean cultivars currently grown in southern Alberta. The base colour of seed coat of AAC PT600 was light brown, similar to AAC Expedition, and was better than Island which had a dark brown base seed coat. The canning and cooking quality attributes of AAC PT600 were similar to the check cultivar Island. AAC PT600 is well suited for commercial production under irrigation in Alberta and Saskatchewan.The presentation of the authors' names and (or) special characters in the title of the pdf file of the accepted manuscript may differ slightly from what is displayed on the item page. The information in the pdf file of the accepted manuscript reflects the original submission by the author
Focal Irradiation Regulates Distal Neural Stem and Progenitor Cell Behaviour
Full text linkNeural stem cells (NSCs) are located along the neuraxis of the adult central nervous system: in the dentate gyrus and subventricular zone (SVZ) of the forebrain as well as the periventricular zone of the spinal cord. Cranial irradiation (IR), a common cancer treatment, leads to fewer brain NSCs and impaired neurogenesis. Herein, we investigated how focal IR impacts distal, non-irradiated NSC niches, termed the “long-distance effect”. We demonstrated that focal cranial IR reduced the numbers of spinal cord derived NSCs at 2 days post-IR. Reciprocally, spinal cord IR led to fewer NSCs and depleted neuroblast pools in the forebrain. Since microglia activation is associated with impaired neurogenesis, we hypothesized that microglia ablation would rescue the IR-mediated effects on forebrain neurogenesis. However, microglia ablation paradigms did not reverse the neural stem and progenitor cell loss. Hence, focal IR alters NSC behaviour along the neuraxis, in the presence or absence of microglia.M.A.S
CES1 Increases Hepatic Triacylglycerol Synthesis Through Activation of PPARγ, LXR and SREBP1c
No full textIncreased hepatic triacylglycerol (TG) storage in lipid droplets (LDs) is a hallmark of metabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction-associated steatohepatitis (MASH). Human carboxylesterase 1 (CES1) regulates TG storage and secretion in hepatocytes, but the mechanism remains to be elucidated. We performed studies in rat hepatoma McArdle RH7777 cells stably transfected with CES1 cDNA and in Ces1d-deficient mice using a variety of biochemical, pharmacological and cell biology approaches including the assessment of gene expression, confocal immunofluorescence microscopy, lipid synthesis measurements and quantitative mass spectrometry. CES1-expressing cells accrued more TG compared to cells lacking CES1 when incubated with oleic acid. CES1 increased the expression of Srebf1c, Nr1h3 and Nr1h2 encoding transcription factors (SREBP1c and LXRα and LXRβ, respectively) that regulate the expression of lipogenic genes. Additionally, CES1 increased the expression of Acsl1 encoding an enzyme catalyzing fatty acid activation and the expression of Dgat1 and Dgat2 encoding enzymes catalyzing TG synthesis. Treatment of CES1-expressing cells with PPARγ antagonist (GW9662), LXR antagonist (GSK2033) or CYP27A1 inhibitor Felodipine prevented CES1-mediated fatty acid esterification into TG. Ces1d-deficient mice fed high-fat diet (HFD) presented with decreased expression of Nr1h3, Nr1h2, Srebf1c and reduced hepatic TG content. Felodipine and GSK2033 treatment eliminated the differential effects on TG concentration between wild-type and Ces1d-deficient hepatocytes. The results suggest that CES1/Ces1d activates PPARγ, LXR and SREBP1c pathways, thereby increasing TG synthesis and LD storage by augmenting fatty acid esterification
Polygenic Risk Scores and Intermediate Phenotypes in Youth Bipolar Disorder
Full text linkBipolar disorder (BD) is a polygenic condition with a heritability of up to 85%. Polygenic risk score(s) (PRS), calculated as the weighted sum of single nucleotide polymorphism risk alleles identified by genome-wide association studies, can be studied as an index of an individual’s genetic liability for BD. Recent studies have found that PRS for BD (BD-PRS) is elevated in adults with BD and youth at familial risk for BD. There is also well-established evidence of neuroimaging and neurocognitive anomalies in youth with BD. Nonetheless, there is ongoing conjecture regarding the validity of BD in youth. One approach to validate youth BD is to demonstrate that it has similar genetic underpinnings as adult BD. This thesis therefore focuses, for the first time, on the polygenic underpinning of BD diagnosis and related intermediate phenotypes in youth with BD. Participants were youth of European ancestry, aged between 13-20 years. DNA was extracted from saliva and genotyped. PRS, based on independent adult genome-wide association summary statistics, were constructed using PRS-CS-auto. Clinical diagnoses were determined by semi-structured interviews. Brain images were collected using 3-Tesla magnetic resonance imaging. Neurocognition was examined using a computerized automated test battery. BD-PRS was significantly higher in youth with BD than controls, with youth at high risk for BD numerically intermediate. Higher BD-PRS was associated with smaller grey matter structure in frontal and temporal regions, poorer white matter integrity in projection tracts, altered resting-state functional connectivity of salience, frontoparietal, and visual network, and poorer neurocognitive performance on sustained attention, decision-making, and affective processing.
In conclusion, the alignment of present youth BD findings with prior adult BD findings adds to the biological validation of BD in youth. This may help reduce doubt and stigma. This cross-sectional research will inform future longitudinal studies, which are needed to elucidate the mechanistic effects of BD-PRS on BD diagnosis and related intermediate phenotypes. Ultimately, continued progress on this line of research has the potential to inform clinical decisions related to early detection, diagnostic classification, risk prediction, and treatment of youth BD.Ph.D
Identifying Plasma Biomarkers That Predict Patient-Reported Outcomes Following Treatment for Trapeziometacarpal Osteoarthritis Using Machine Learning
Full text linkOsteoarthritis (OA) in the trapeziometacarpal joint (TM) is a prevalent form of hand OA, yet research on biomarkers specific to hand OA remains limited. This study aims to identify systemic plasma biomarkers at baseline in TM OA patients that are associated with patient-reported outcomes one year post-treatment. Blood samples and clinical data were collected prospectively from 143 TM OA patients undergoing conservative therapy, fat grafting, or surgery, with one-year follow-up. Supervised machine learning with Lasso regularization analyzed associations among 10 systemic biomarkers related to cartilage turnover, bone remodeling, pain, or lipid metabolism. Generalized estimating equation models evaluated baseline biomarker associations with one-year outcomes. Patients averaged 61 years, were mostly female (69%), and were primarily treated conservatively (47%). The machine learning model identified associations between outcomes and biomarkers, including PIIANP, Visfatin, adiponectin, and leptin. Adjusted analyses revealed baseline PIIANP associated with VAS, QuickDASH, and TASD improvements, while Visfatin correlated with VAS worsening. We could identify two different plasma markers that could predict the clinical outcome of TM OA treatment. Baseline PIIANP is associated with improvement, while Visfatin is associated with worsening in TM OA outcomes up to one year post-treatment. Further prospective studies are needed to confirm and solidify these findings
A Systematic Review of Aircraft Disinsection Efficacy
Full text linkDisinsection of aircraft and other conveyances is recommended to prevent the international spread of disease-carrying mosquitoes. A systematic review synthesized the effectiveness of adult mosquito disinsection aboard international air, marine, and land conveyances, using literature available up to 31 May 2025. Nineteen experimental trials, nine of which included an unexposed control arm, were synthesized. The studies were generally of poor quality with high risk of bias, and adherence to WHO guidelines was 33.30% (range: 18.20–60.5%). Across comparator trials of aircraft disinsection, the odds of mosquito mortality in the treated groups compared to control groups was 163.60 (95% CI 147–182), and the relative risk of mosquito death was 14.24 (95% CI 12.99–15.63). The direction of effect was consistent across mosquito species, methods of disinsection, types of aircraft, and insecticides, though the magnitude of the effect varied widely. The only WHO-recommended insecticide tested in a controlled trial was 2% d-phenothrin, which demonstrated an odds ratio of 171.70 (95% CI 139.10–212) and a relative risk of 20.08 (95% CI 16.53–24.43) for mosquito mortality. The lack of adherence to WHO guidelines raises uncertainty about the real-world effectiveness of disinsection