Phenol is an important chemical engineering material and ubiquitous in industry wastewater,its existence has become a thorny issue in many developed and developing country.More and more stringent standards for effluent all over the world with human realizing the toxicity of phenol have been announced.Many advanced biological methods are applied to industrial wastewater treatment with low cost,high efficiency and no secondary pollution,but the screening of function microorganisms is certain cumbersome process.In our study a rapid procedure devised for screening bacteria on solid medium can degrade phenol coupled with attenuated total reflection fourier transform infrared(ATR-FTIR) which is a detection method has the characteristics of efficient,fast,high fingerprint were used.Principal component analysis(PCA) is a method in common use to extract fingerprint peaks effectively,it couples with partial least squares(PLS) statistical method could establish a credible model.The model we created using PCA-PLS can reach 99.5% of coefficient determination and validation data get 99.4%,which shows the promising fitness and forecasting of the model.The high fitting model is used for predicting the concentration of phenol at solid medium where the bacteria were grown.The highly consistent result of two screening methods,solid cultural with ATR-FTIR detected and traditional liquid cultural detected by GC methods,suggests the former can rapid isolate the bacteria which can degrade substrates as well as traditional cumbersome liquid cultural method.Many hazardous substrates widely existed in industry wastewater,most of them has specialize fingerprint peaks detected by ATR-FTIR,thereby this detected method could be used as a rapid detection for isolation of functional microorganisms those can degrade many other toxic substrates.