Broadband generation of multiple high-order OAM modes in ring-core fibers using multi-pitch chirped long-period fiber gratings.

A multi-order broadband mode converter in a ring-core fiber (RCF) using a multi-pitch chirped long-period fiber grating (LPFG), where multiple pitches were introduced in each chirp to further increase the bandwidth, is proposed and demonstrated. The grating parameters were optimized both theoretically and experimentally to achieve broadband mode generation of OAM ± 2 and OAM ± 3 modes by increasing the number of chirps and pitches. The mode conversion efficiency is higher than 90% with a broadband of 57 nm from 1456 nm to 1513 nm and 51 nm from 1573 nm to 1624 nm, corresponding to the second-order OAM mode and third-order OAM mode, respectively. Additionally, the insertion loss is less than 0.8 dB, and the purity is over 90%. The demonstrated mode converter has successfully achieved simultaneous generation of multi-order broadband OAM modes in a RCF for the first time, which has promising potential for application in OAM mode-division multiplexing systems.

Ultra-low-loss 5-LP mode selective coupler based on fused biconical taper technique.

Trapped in the stringent adiabatic transmission condition of high-order modes, low-loss fused biconical taper mode selective coupler (FBT-MSC) has long been challenging to achieve. We identify the adiabatic predicament of high-order modes to stem from the rapid variation of the eigenmode field diameter, which is caused by the large core-cladding diameter difference of few-mode fiber (FMF). We demonstrate that introducing a positive-index inner cladding in FMF is an effective approach to address this predicament. The optimized FMF can be used as dedicated fiber for FBT-MSC fabrication, and exhibits good compatibility with the original fibers, which is critical for the wide adoption of MSC. As an example, we add inner cladding in a step-index FMF to achieve excellent adiabatic high-order mode characteristics. The optimized fiber is used to manufacture ultra-low-loss 5-LP MSC. The insertion losses of the fabricated LP01, LP11, LP21, LP02 and LP12 MSCs are 0.13 dB at 1541 nm, 0.02 dB at 1553 nm, 0.08 dB at 1538 nm, 0.20 dB at 1523 nm, and 0.15 dB at 1539 nm, respectively, with smoothly varying insertion loss across the wavelength domain. Additional loss is less than 0.20 dB from 1465.00 nm to 1639.31 nm, and the 90% conversion bandwidth exceeds 68.03 nm, 166.68 nm, 174.31 nm, 132.83 nm, and 84.17 nm, respectively. MSCs are manufactured using commercial equipment and a standardized process that takes just 15 minutes, making them a potential candidate for low-cost batch manufacturing in a space division multiplexing system.

First Report of Stemphylium lycopersici Causing Leaf Spot on Hevea brasiliensis in China.

Hevea brasiliensis is widely planted in tropical and subtropical regions and is the main source of natural rubber production. The growth of rubber trees is plagued by various leaf diseases, resulting in decreased rubber production. From January to March in 2020, a severe leaf spots disease on Hevea brasiliensis found in Agricultural Science Base in Haidian campus of Hainan University (20° 03' 31″ N, 110° 19' 07″ E), Haikou, Hainan province, China. Spots were only observed on the mature green rather than young and bronze-colored leaves. This symptom has never been reported on the leaves of Hevea brasiliensis. During the early stages of the disease, gray leaf spots were concentrated to the leaf margins, but later expanded forming irregular gray lesions with chlorotic edges (Figure 1A). Eventually, lesions became necrotic shot holed, and leaves curled, wilted, and dropped. Five small pieces were cut from the margin of spots from different infected leaves, and were surface disinfected with 75% alcohol three times for five seconds each time and 1% sodium hypochlorite solution (NaClO) for 60 s. After washing twice with sterile water, leaf pieces were placed in the center of plates with Potato Dextrose Agar (PDA) medium and incubated for one week at 28 °C. After 7 days, mycelium developed and colonies were single-spore cultured for further study. One of the strains labeled HN01 developed a yellowish-brown to reddish-brown pigment on PDA, and the colonies were gray and cottony. The colony and pigment feature very consistent with Stemphylium sp. (Figure 2) (Li et al. 2017). Conidiophore were solitary, transparent to pale, mostly 102.1-228.8 µm × 4.0-5.8 µm, with 2-3 septa and apical vesicular swellings 6.5-7.9 µm. The dimensions of conidia were 28.3-45.1 × 11.5-17.5 µm and one septum (Figure 3). Conidia of S. lycopersici were solitary, oblong with a conical end at the apex, with 1-2 septa, and constricted at the transverse septum. The internal transcribed spacer region of rDNA was amplified with primers ITS1/ITS4 (5'-TCCGTAGGTGAACCTGCGG-3'/5'-TCCTCCGCTTATTGATATGC-3'), glyceraldehyde-3-phosphate dehydrogenase (gpd) was amplified with primers GPD-F/R (5'-GCACCGACCACAAAAATC-3'/ 5'-GGGCCGTCAACGACCTTC-3'), calmodulin region (cmdA) was amplified with the primers CALDF1/CALDR2 (5'-AGCAAGTCTCCGAGTTCAAGG-3'/5'-CTTCTGCATCATCAYCTGGACG3') from genomic DNA of strain HN01 (Xie et al. 2018), and PCR products were sequenced. The ITS sequence of strain HN01 (GenBank Accession No. MZ496930) had 99.64% identity with isolates sl001, sl110, sl111, and sl112 of Stemphylium lycopersici (GenBank Accession No. KX858848.1, MF480547.1, MF480548.1, MF480549.1). Similarly GPD sequences (GenBank Accession No. MZ505106) had 100% identity with strain xiqing, HZ2114 and HZ2115 of Stemphylium lycopersici (GenBank Accession No. KR911809.1, KR911810.1, KT957742.1 and KT957743.1), and CMDA sequences (GenBank Accession No. MZ505105) had 99.85% identity with Stemphylium lycopersici strain LJ1609270201 (GenBank Accession No. MG742412.1). A phylogenetic analysis constructed by MEGA6.0 based on concatenated sequences of the HN01 and another 17 strains from GenBank by using the maximum-likelihood (ML) method showed that the HN01 was clustered and matched with Stemphylium lycopersici LJ1609270201 (Figure 4). To satisfy Koch's postulates, we inoculated mature green leaves of Hevea brasiliensis with mycelial plugs (diameter = 5 mm) of pure cultured strain HN01. All leaves of Hevea brasiliensis were wrapped in a freezer bag to maintain relative humidity >85%, and the temperature of greenhouse is 28ºC. The disease developed on the inoculated leaves after 2-3 days, but not on control leaves (Figure 1B). We used the same method as before to re-isolate the pathogen, which had the same morphology and genotypes as the original isolate. S. lycopersici has been reported to infect the leaves of a variety of plants, including pepper, tomato, eggplant, watermelon, Physalis alkekengi. (Yang et al.2017; Ben et al. 2017; Yang et al. 2020). To our knowledge, this is the first record of S. lycopersici causing leaf spot of Hevea brasiliensis in China, and Hevea brasiliensis is the global new host of S. lycopersici. Hevea brasiliensis is the main source of natural rubber and is widely planted in southern China. Therefore, it is imperative to implement disease management measures to prevent potential threats.

Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmark.

Federated learning has emerged as a promising paradigm for privacy-preserving collaboration among different parties. Recently, with the popularity of federated learning, an influx of approaches have delivered towards different realistic challenges. In this survey, we provide a systematic overview of the important and recent developments of research on federated learning. Firstly, we introduce the study history and terminology definition of this area. Then, we comprehensively review three basic lines of research: generalization, robustness, and fairness, by introducing their respective background concepts, task settings, and main challenges. We also offer a detailed overview of representative literature on both methods and datasets. We further benchmark the reviewed methods on several well-known datasets. Finally, we point out several open issues in this field and suggest opportunities for further research. We also provide a public website to continuously track developments in this fast advancing field: https://github.com/WenkeHuang/MarsFL.

Generalizable Heterogeneous Federated Cross-Correlation and Instance Similarity Learning.

Federated learning is an important privacy-preserving multi-party learning paradigm, involving collaborative learning with others and local updating on private data. Model heterogeneity and catastrophic forgetting are two crucial challenges, which greatly limit the applicability and generalizability. This paper presents a novel FCCL+, federated correlation and similarity learning with non-target distillation, facilitating the both intra-domain discriminability and inter-domain generalization. For heterogeneity issue, we leverage irrelevant unlabeled public data for communication between the heterogeneous participants. We construct cross-correlation matrix and align instance similarity distribution on both logits and feature levels, which effectively overcomes the communication barrier and improves the generalizable ability. For catastrophic forgetting in local updating stage, FCCL+ introduces Federated Non Target Distillation, which retains inter-domain knowledge while avoiding the optimization conflict issue, fulling distilling privileged inter-domain information through depicting posterior classes relation. Considering that there is no standard benchmark for evaluating existing heterogeneous federated learning under the same setting, we present a comprehensive benchmark with extensive representative methods under four domain shift scenarios, supporting both heterogeneous and homogeneous federated settings. Empirical results demonstrate the superiority of our method and the efficiency of modules on various scenarios. The benchmark code for reproducing our results is available at https://github.com/WenkeHuang/FCCL.

Effective adsorption and catalytic reduction of nitrophenols by amino-rich Cu(I)-I coordination polymer.

Nitrophenols are identified as the priority organic pollutants due to the chemical stability, water solubility, persistence, and toxicity to human health and the environment. Hence, removal of nitrophenols from waste water is vitally essential. In this study, amino-rich coordination polymer Cu2I2(MA)2 (MA = melamine) has been applied for efficient adsorption and catalytic reduction of nitrophenols, like 4-nitrophenol (4-NP), 2, 4-dinitrophenol (DNP) and 2, 4, 6-trinitrophenol (TNP). The effect of various parameters like contact time, initial concentrations, pH, and temperature on adsorption were investigated. The adsorption of nitrophenols fitted the pseudo-second-order kinetic model and Langmuir isotherms model well. The maximum adsorption capacities were 285.71, 232.02, and 131.57 mg g-1 for 4-NP, DNP, and TNP when initial concentrations were 50 mg L-1 at 293.15 K, respectively. The adsorption of nitrophenols is a spontaneous, endothermic, and entropy-driven process. The reduction reaction followed the pseudo-first-order kinetics, and the kinetic rate constants were 0.4413, 0.3167, and 0.17538 min-1 for 4-NP, DNP, and TNP, respectively. The effect of initial nitrophenols concentration, anions, and temperature on reduction process was investigated. The mechanism of adsorption and catalytic reduction of Cu2I2(MA)2 was studied. The results demonstrated that Cu2I2(MA)2 exhibits excellent adsorption and catalytic activity to remove nitrophenols.

Characterization of conidial autofluorescence in powdery mildew.

Autofluorescence is produced by endogenous fluorophores, such as NAD(P)H, lipofuscin, melanin, and riboflavin, indicating the accumulation of substances and the state of energy metabolism in organisms. As an obligate parasite, powdery mildew is wildly spread by air and parasitic crops. However, most identification studies have been based on morphology and molecular biology which were far too time- and labor-consuming, thus lacking characteristic, simple, and effective means. Using microscopy under the blue and cyan channels, we elaborated visible conidial autofluorescence in three powdery mildew species, Erysiphe quercicola, E. cichoracearum, and Podosphaera hibiscicola, with a sharp increase during the conidia senescence in E. quercicola. Additionally, the main spectral excitation detected by fluorescence spectrometery was 375 nm for these species, with a common emission peak at approximately 458-463 nm, and an additional trend at 487 nm for P. hibiscicola. Because NAD(P)H has a similar spectral feature, we further investigated the relation between NAD(P)H and conidial autofluorescence by fluorescence spectra. We observed that the reduced coenzymes prominently contributed to conidial autofluorescence; however, the conidial autofluorescence in P. hibiscicola displayed a different trend that may be affected by the oxidized coenzyme -NAD. Finally, the normalized average spectra of these three powdery mildew species and standard samples showed that the spectral trend of each species was similar but that the features in detail were specific and distinct based on principal component analysis. In conclusion, we showed and characterized conidial autofluorescence in three powdery mildew species for the first time. The specific conidial autofluorescence in these species provides a new idea for the development of field spore capture and identification devices for the discrimination of powdery mildew at the species level.