塑料回收的光谱学

Background

The synthetic materials comprising some plastics use macromolecules such as polymers. These polymers are commonly manufactured through injection molding at low temperatures, suggesting that reuse of the materials via recycling is simple. However,非常少数类型的塑料可以混合到均匀的混合物中，这使得有必要在再循环流中干净地分离不同的聚合物。

因为NIR光谱和拉曼分析等技术提供了有关包括化学成分的材料特性的有用信息，它们在识别回收塑料方面是有效的。光谱学到再循环流的集成可以通过减轻对某些分类步骤的需要更简单。

在NIR波长（> 800nm）处，聚合物具有强大，不同的光谱特征，其与印在塑料上印刷的回收码密切相关（图1）。此外，通过NIR光谱学，还有很少的样品制剂进行。设置灵活，可定制到许多不同的情况。

Figure 1.米ost coded plastics have distinct spectral features in the NIR. Spectrometers can be integrated into recycling processes to help sort plastic types.

In some cases, Raman spectroscopy can be used to identify unknown compounds. Indeed, under the right conditions – i.e., at sufficient signal levels and with adequate methods to mitigate interference from fluorescence -- Raman is highly specific for plastics identification.

与拉曼和NIR光谱一样有效，可以用于透明和彩色塑料鉴定，既不适用于鉴定黑色聚合物。这是因为炭黑吸收了所有拉曼励磁激光和NIR区域中的所有光。然而，文献中有用于使用用于识别黑色聚合物的中红外光谱技术的文献。¹

Diffuse Reflectance with Flame-NIR+ Spectrometer

出于本申请说明的目的，我们的重点是在NIR光谱上。为了证明塑料鉴定的燃烧+光谱仪的有效性，我们测量了几种样品的漫反射率：Pete（聚对苯二甲酸乙二醇酯），通常用于容器和衣服;热塑性LDPE（低密度聚乙烯）;PP（聚丙烯），其日常应用众多;和PS（聚苯乙烯），这使其进入各种包装和容器。

The火焰+(Figure 2) has a compact, high-performance optical bench with an uncooled InGaAs array detector for spectral response from 970-1700 nm. It has up to 6x increased sensitivity compared with earlier models, offering lower limits of detection and allowing for shorter integration times.

Figure 2.凭借其紧凑的占地面积和低功耗需求，Flame-Nir +是一项有吸引力的选择，可以集成到回收流中。

To measure our samples, we used the Flame-NIR+ with a high-power tungsten halogen light source, 600 µm reflection probe and diffuse reflectance standard. OceanView operating software completed the setup. We set our parameters at 6 ms integration time, with 50 scans to average and boxcar width set to 1.

OceanView acquired data in absorbance mode, so we could plot log(1/R) of the reflectance spectra, withRbeing the reflectance of the sample. This can be a more intuitive way to show reflectance spectra for chemical analysis, as taking log(1/R) of the reflectance spectrum allows us to see concentration as scaling with signal intensity.

Initally, we collected Flame-NIR+ spectra without applying baseline correction (Figure 3). To make it easier to discern spectral features, we did a baseline correction by subtracting the reflectance at 1303 nm from every data point in the spectrum. This is a data processing step that helps to eliminate the intensity offset that can result from fibers moving or other issues, making it easier to examine differences in spectral shapes (Figure 4).

Figure 3.Diffuse reflectance of plastic samples using the Flame-NIR+ spectrometer (970-1700 nm). Data is shown without baseline correction.

图4。With baseline correction applied, diffuse reflectance spectra measured with the Flame-NIR+ spectrometer show clear differences among plastic types.

漫反射率与NIRQuest +光谱仪

Additional measurements were performed using thenirquest + 1.7, a thermoelectrically cooled InGaAs-array detector spectrometer with response from 900-1700 nm (Figure 5). Versions are also available with response from 900-2200 nm and from 900-2500 nm.

Figure 5.Nirquest + 1.7光谱仪（900-1700 nm）是一种具有极大的热稳定性的高灵敏度仪器，这使得工艺环境是一个很好的选择。

Nirquest +光谱仪具有增强的光学台面设计，可用于高灵敏度性能，允许在较长波长下漫反射测量的检测和清洁光谱的低限制。与Flame-Nir +相比，NirQuest +更敏感，并提供频谱响应的版本，较长的NIR波长（> 1700nm）。

与Flame-nir +一样，我们使用高功率钨卤素光源，600μm反射探针和漫反射标准测量塑料样品（PET和PP）。我们在5毫秒的集成时间下将参数设置为5毫秒，平均扫描到平均和箱式宽度设置为1.我们将结果缩放为％反射（图6）。

图6。Diffuse reflectance of polypropylene measured with the NIRQuest+1.7 spectrometer.

米aking Use of NIR Data

使用NIR Spectra来构建强大的校准模型，但如果您试图在这些校准模型之外预测，则它们不是很可行。实际上，NIR Spectra提供了有关您样本的丰富信息，只要您有能够提取该数据的工具。例如，主成分分析是一种化学计量分析方法，可以应用于收集到组的数据，并使用它们的光谱签名分类样本。

此外，对于塑料分拣过程，能够与您的NIR反射光谱进行有用的决策是我们的绝佳机会Ocean Intelligence机器学习算法。应用于自动回收分选过程中，海洋智能提供了另一种洞察层，在更有效地分离聚合物并且具有较少的分类误差。任何没有正确排序的东西都变得浪费，这必须再次回收。

结论

随着NIR光谱仪和机器学习专业知识的持续演变，回收商可以简化塑料分拣过程，使操作更具成本效益，并有助于更可持续的环境。

下载此应用程序注意的PDF。

¹Becker, W., Sachsenheimer, K. And Klemenz, M., “Detection of Black Plastics in the Middle Infrared Spectrum (MIR) Using Photon Up-Conversion Technique for Polymer Recycling Purposes,“聚合物(Basel). 2017 Sep; 9(9): 435.