Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1945
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-19T22:03:21.978274Z
status: published
lens: g_space
sector: climate
headline: AI-Enabled Mixed Potential Sensor Array for Complex VOC Mixture Decoding and Real-Time Methanol-Risk Assessment in Baijiu
dek: Accurate identification of complex gas mixtures in dynamic real-world environments is essential for environmental monitoring, food safety, and public health. However, these challenges arise mainly from overlapping response fingerprints of chemically similar hazardous analytes and the limited capability of conventional sensors to decouple multicomponent, concentration-coupled signals. Here, we report a ZnFe2O4/MoO3 mixed potential sensor array enabled by a material-device-algorithm co-design strategy. Through com…
gain_title: A ZnFe2O4/MoO3 sensor array paired with a CNN-BiLSTM model separated methanol, ethanol and n-butanol mixtures and identified Baijiu samples over the statutory methanol limit with above 98.89% accuracy.
problem_title: (none)
trace_subject: (none)
gain_reading: A ZnFe2O4/MoO3 sensor array paired with a CNN-BiLSTM model separated methanol, ethanol and n-butanol mixtures and identified Baijiu samples over the statutory methanol limit with above 98.89% accuracy.
gain_evidence: achieving above 98.89% accuracy in identifying samples exceeding the statutory limit
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built a mixed potential sensor array from ZnFe2O4 and MoO3 electrodes that produces distinct response fingerprints for six volatile organic compounds. A CNN-BiLSTM deep-learning model reads the dynamic signals to classify alcohol mixtures and predict their concentrations at the same time.

The approach was extended to a portable platform for assessing methanol risk in Baijiu, a complex liquor matrix. This matters because chemically similar hazardous compounds are hard for conventional sensors to separate. The report describes a deployable route, but the abstract gives no evidence of performance beyond the tested samples and settings.
limitation: 
tag: Evidence-backed gain
key_points: The array gave response fluctuation standard deviations of 0.47-2.43 mV across all analytes. | Response time was 12 s toward 100 ppm triethylamine. | The model reached 100% accuracy on the test set and 99.33% on an independent validation set. | Mean absolute errors for concentration prediction were 1.61 and 1.62 ppm on the two sets.
rundown: The sensing design uses compositional engineering of the electrodes so the array yields orthogonal, cross-reactive fingerprints. This targets the overlapping responses of chemically similar hazardous analytes that limit conventional sensors.

For alcohol mixtures with strong competitive adsorption, namely methanol, ethanol and n-butanol, the CNN-BiLSTM spatiotemporal fusion model extracts multiscale features from the dynamic sensing responses. It performs mixture classification and concentration prediction simultaneously.

The authors then built a portable Baijiu methanol sensing platform for real-time risk assessment, identifying samples exceeding the statutory limit with above 98.89% accuracy.
sources:
- peer_reviewed | ACS Sensors | https://doi.org/10.1021/acssensors.6c02752 | 2026-10-16
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

bb11ce007d6f4c3b130f1a344b91090b3c4135c8f90bfb8e1289f3a091514162

previous

0000000000000000000000000000000000000000000000000000000000000000