Ep 35 — Agriculture and Soil: Organic Matter Content and Fertilizer Analysis

Series: Infrared Spectroscopy Encyclopedia: From Principles to Practice
Chapter: Part 3 · Intermediate — Industry Applications (Later segment, final episode of Intermediate)
Target Audience: Graduate students in soil science/agronomy/plant nutrition, agricultural testing technicians, precision agriculture engineers, fertilizer quality control personnel
Prerequisites: Ep 06 (C-H, C=O), Ep 13 (Transmission), Ep 14 (ATR), Ep 19 (Library Search), Ep 20 (Quantitative Analysis), Ep 27 (Environmental Soil Testing)
Reading Time: Approximately 45 minutes

Introduction: How Much Chemical Information Can Be Hidden in a Soil Sample?

In 2008, the team of Janik and Skjemstad from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) published a study in Soil Biology & Biochemistry that is still highly cited. They used mid-infrared DRIFTS (Diffuse Reflectance Fourier Transform Infrared Spectroscopy) + Partial Least Squares Regression (PLSR) to measure the organic carbon content in over 1500 Australian soil samples, achieving a correlation coefficient R² of 0.93 with the dichromate oxidation method (Walkley-Black method). The measurement speed was reduced from 4 hours per sample to 1 minute [1].

"Mid-infrared diffuse reflectance spectroscopy, combined with chemometric modeling, has emerged as a revolutionary tool for soil analysis, replacing time-consuming wet chemistry for high-throughput soil characterization."
—— Janik L J, Skjemstad J O. Soil Biology & Biochemistry 2008 [1]

A similar revolution is occurring simultaneously in the field of fertilizer and crop quality analysis. Today, near-infrared spectrometers are standard equipment in large grain depots in Europe and America—a truckload of wheat enters the depot, and within 30 seconds, three key indicators (protein, moisture, wet gluten content) are provided [2]. In this episode, we will systematically discuss the applications of FTIR/NIR in agriculture and soil: from soil organic matter assessment to fertilizer composition identification, from crop quality rapid testing to practical applications in precision agriculture, and finally summarize all key points of industry applications in the intermediate chapter.

1. Rapid Assessment of Soil Organic Matter (SOM) by Infrared Spectroscopy

1.1 What is Soil Organic Matter?

Soil Organic Matter (SOM) refers to the total carbon-containing organic substances in soil, including [1][3][4]:

• Humic substances: humic acid, fulvic acid, humin, accounting for 60–80% of SOM;
• Undecomposed/semi-decomposed plant residues: 10–20%;
• Microbial biomass: 1–5%;
• Dissolved organic carbon (DOC): < 1%.

SOM content is a core indicator of soil fertility—it underpins water retention, nutrient retention, structural stability, and microbial activity [3]. Traditional measurement methods (Walkley-Black dichromate oxidation method, CNS elemental analysis) are time-consuming, reagent-intensive, and produce waste, making them unsuitable for high-throughput detection [1][3].

1.2 Infrared Characteristics of SOM

Infrared characteristic peaks of SOM [3][4][5]:

Table 1: Mid-infrared characteristic peaks of soil organic matter (data sources: Reeves Appl Spectrosc Rev 2010 [3]; Demyan Soil Biol Biochem 2012 [4])

🔗 Further Reading: The aliphatic CH₂ vibrations (2925/2850 cm⁻¹) in SOM are key to assessing the "lipidification degree" of organic matter. See ftir.fun alkyl C-H functional group page. The intensity ratio of 2925/2850 cm⁻¹ reflects aliphatic carbon content; 1720 cm⁻¹ reflects carboxylic acid/ester content; their ratio can serve as an indicator of SOM "humification degree" [4].

1.3 Mid-Infrared vs Near-Infrared: Which is Better for SOM?

Table 2: Comparison of mid-infrared and near-infrared for SOM analysis (data source: Nocita Adv Agron 2015 [5])

Practical experience [1][5]:

• Research, accurate models: MIR-DRIFTS (richer chemical information);
• Field rapid screening: Portable NIR (fast, portable);
• Routine laboratory: Combine MIR and NIR for cross-validation.

1.4 Sample Preparation for DRIFTS Measurement of SOM

DRIFTS (Diffuse Reflectance) is the primary method for MIR measurement of soil [3][5]:

1. Air-drying: Soil samples dried at 40°C for 24 h (eliminate water interference; see Section 6);
2. Grinding: Grind with an agate mortar to < 80 mesh (< 0.18 mm);
3. Dilution: Take 5 mg soil + 200 mg KBr (1:40), mix and grind for 2 minutes;
4. Loading: Fill the DRIFTS sample cup, level the surface;
5. Measurement: Scan 64 times, resolution 4 cm⁻¹, background with KBr powder;
6. Kubelka-Munk transformation: Convert reflectance into equivalent absorbance spectrum.

💡 Industry Tip: Soil must be thoroughly dried and ground. Water causes strong absorption at 3400/1640 cm⁻¹, masking SOM information; coarse particles lead to severe scattering and baseline tilt [3][5].

1.5 Chemometric Modeling

SOM content cannot be read from a single peak; multivariate regression is required [1][3][5]:

• Preprocessing: SNV (Standard Normal Variate) + first derivative (remove baseline drift);
• Modeling: PLSR (Partial Least Squares Regression), 10-fold cross-validation;
• Variable selection: CARS (Competitive Adaptive Reweighted Sampling) to select key wavenumbers;
• Model evaluation: R², RMSE, RPD (Residual Predictive Deviation):
  • RPD < 2: Model unreliable;
  • RPD 2–3: Usable;
  • RPD > 3: Excellent;
  • RPD > 5: Can replace laboratory method.

The Janik team's SOM model achieved R² = 0.93, RPD = 3.6, reaching the "excellent" level [1].

2. Fertilizer Composition Identification and Adulteration Detection

2.1 The Adulteration Problem in the Fertilizer Market

Fertilizer is the "food" of modern agriculture. According to the International Fertilizer Association (IFA) data [6][7]:

• Global annual fertilizer consumption > 200 million tons;
• Fertilizer adulteration rate in some developing countries as high as 20–30%;
• Main adulteration methods:
  • Filling high-value fertilizers (e.g., KCl) with cheap salts (NaCl, Na₂SO₄);
  • Substituting urea for ammonium sulfate (similar nitrogen content, price difference 3×);
  • Replacing diammonium phosphate with industrial waste salts;
  • Adding colorants to mask compositional differences.

Traditional chemical testing (Kjeldahl method, molybdenum blue method, flame photometry) is time-consuming and requires professionals [6]. ATR-FTIR and portable NIR are playing increasingly important roles in rapid fertilizer testing [6][7].

2.2 Infrared Characteristics of Common Fertilizers

Table 3: Infrared characteristics of common fertilizers (data sources: Garcia Talanta 2019 [6]; FAO Fertilizer Manual [7])

2.3 Adulteration Identification Strategies

ATR-FTIR combined with chemometrics has become a standard method for adulteration identification [6][8]:

1. Library building: Collect 50+ authentic fertilizer samples to establish a reference library;
2. Scanning: Place a small amount of fertilizer granules on diamond ATR, obtain spectrum in 30 seconds;
3. Library search: Compare HQI with reference library; > 0.95 judged as authentic;
4. Chemometric discrimination: PCA/PLS-DA models determine adulteration;
5. Quantitative analysis: PLS models predict N/P/K content.

Typical adulteration cases [6][8]:

• NaCl adulterated in urea: No Cl⁻ absorption in authentic products; adulterated samples show weak water peaks at 1640/1380 cm⁻¹ (NaCl hygroscopic);
• (NH₄)₂CO₃ adulterated in ammonium sulfate: CO₃²⁻ impurity peak at 1430 cm⁻¹;
• Gypsum adulterated in DAP: impurity peaks at 1430 cm⁻¹ and 1110 cm⁻¹ (SO₄²⁻);
• Formulation fraud in compound fertilizers: relative intensities of characteristic peaks inconsistent with formulation.

2.4 Case Study: Urea Adulteration Screening in the Indian Market

A 2019 study commissioned by the Indian Ministry of Agriculture and conducted by IICT (Indian Institute of Chemical Technology) [6]:

• Samples: 320 commercial urea samples from 5 Indian states;
• Method: ATR-FTIR + PLS-DA;
• Results:
  • 76 samples found adulterated (24% adulteration rate);
  • Main adulterants: NaCl (46 cases), (NH₄)₂SO₄ (20 cases), sawdust (10 cases);
  • Model accuracy 96% (validation set of 80 samples);
  • Measurement time: < 1 minute per sample.

The entire screening process was more than 50 times faster than traditional methods, providing key technology for the Indian government to regulate the fertilizer market [6].

3. Near-Infrared Rapid Testing of Crop Quality

3.1 Why is NIR Suitable for Grain Analysis?

Near-infrared (NIR) spectroscopy is most maturely applied in agriculture for grain quality analysis [2][9]. The main components of grain:

• Starch (overtones and combinations of C–H, O–H, C–O stretches);
• Protein (N–H overtones);
• Moisture (O–H overtones);
• Fat (C–H overtones).

These components have characteristic absorptions in the NIR region (780–2500 nm, corresponding to 12800–4000 cm⁻¹):

Table 4: Common NIR wavelengths for grain analysis (data source: Williams Near-Infrared Technology [9])

Advantages of NIR [2][9]:

• Almost no sample preparation (whole wheat can be directly measured);
• Simultaneous multi-component analysis within 30 seconds;
• No reagents, no waste liquid;
• Robust instruments suitable for on-site use at grain depots and purchase points.

3.2 Main Applications

Main applications of NIR in crop quality analysis [2][9][10]:

Table 5: Applications of NIR in crop quality analysis (data source: FOSS Infratec Reference [10])

3.3 Practical Inbound Testing at Grain Depots

Inbound process at large European and American grain depots [10]:

1. Sampling: Take 5 kg sample from a 30-ton truck using quartering method;
2. Subsampling: Reduce to 200 g;
3. Whole grain scanning: Use FOSS Infratec 1241 whole-grain NIR, obtain protein, moisture, and test weight in 30 seconds;
4. Pricing: Grade based on protein content, determine purchase price;
5. Inbound classification: Store by quality grade to ensure subsequent processing quality.

The entire process takes less than 2 minutes, handling 1000+ trucks per day, a throughput unattainable by traditional wet chemistry methods [10].

3.4 Progress in China

Domestic application of NIR in agriculture began in the 1990s and has rapidly expanded in recent years [2][11]:

• National grain reserves: NIR standard for inbound quality inspection;
• Large flour mills: Online NIR real-time monitoring of wheat flour ash and moisture;
• Feed mills: NIR testing of raw material nutritional components;
• Seed companies: NIR for early-generation breeding screening;
• Tea enterprises: NIR testing of tea polyphenols and caffeine content.

Domestic instruments (Shanxi Huake, Beijing Kesh, Shanghai Lingguang) have gradually replaced imports, with prices only one-third of imported ones [11].

4. Application of Infrared Spectroscopy in Precision Agriculture

4.1 Core Concepts of Precision Agriculture

Precision agriculture manages inputs according to the "4R" principles [12][13]:

• Right time: Fertilize/irrigate at the correct time;
• Right place: Fertilize/irrigate at the correct location;
• Right rate: Use the correct amount;
• Right source: Use the correct fertilizer type.

Role of infrared spectroscopy in precision agriculture [12][13]:

│ Soil Survey (DRIFTS/NIR Mapping) │
│ → SOM, N, P, K, pH distribution per field │
│ → Determine base fertilizer ratio │
├─────────────────────────────────────┤
│ Crop Growth Monitoring (Leaf NIR Reflectance) │
│ → Real-time nitrogen and chlorophyll status │
│ → Determine topdressing timing and amount │
├─────────────────────────────────────┤
│ Yield Prediction (UAV Hyperspectral) │
│ → Large-area crop growth assessment │
│ → Harvest timing decision │
├─────────────────────────────────────┤
│ Quality Quick Check (Grain NIR) │
│ → Incoming quality grading │
│ → High quality, high price │
└─────────────────────────────────────┘

| Ep 32 | Cultural Heritage Conservation | Pigments, binders, cross-sectional imaging |
| Ep 33 | Biomedical | Cancer diagnosis, bacteria identification, tissue imaging |
| Ep 34 | Semiconductor Electronics | Surface contamination, thin film analysis |
| Ep 35 | Agriculture & Soil | SOM, fertilizers, crop quality, precision agriculture |

7.2 Common Industry Application Experiences

From Ep 21–35 we can extract several common industry experiences:

1. Sampling method selection determines everything: Solid powder → ATR/DRIFTS; Liquid → ATR; Gas → Long-path gas cell; Thin film → GIR/transmission; Micro samples → μ-FTIR; Surface imaging → FPA;
2. Chemometrics is an essential skill: Modern infrared analysis relies on PLSR, PCA, SVM, CNN;
3. Databases are core competitiveness: General libraries (NIST, IRUG) + self-built industry libraries combined;
4. Method standardization is key to translation: ASTM, ISO, industry SOPs are the bridge to industry implementation;
5. Multi-technique complementarity: FTIR + Raman, XPS, MS, ICP-MS, etc., provide a complete evidence chain.

7.3 Recommended Learning Paths

After completing the intermediate section, it is suggested that readers choose according to their industry direction:

• Research direction: Enter Ep 36–45 Advanced section, learn μ-FTIR, SR-FTIR, O-PTIR, 2D-COS, hyphenated techniques, chemometrics, machine learning and other cutting-edge techniques;
• Instrument direction: Enter Ep 46–55 Instruments and Tools section, deeply understand various brand instruments, accessories, maintenance and troubleshooting;
• Data direction: Enter Ep 53–55 Open Source Tools section, learn open-source analysis platforms such as SpectroChemPy, HyperSpy, Photizo.

Summary of This Episode

Discussion Questions

1. You need to measure the SOM content of a batch of soil samples, but find that the water peaks at 3400/1640 cm⁻¹ in the MIR spectrum are strong, masking the carboxylic acid C=O peak at 1720 cm⁻¹. Please design three solutions to address water interference.

2. In fertilizer adulteration identification, what changes occur in the infrared spectrum when NaCl is mixed into urea (NH₂)₂CO? Can FTIR alone quantitatively detect the NaCl doping ratio? Why?

3. Compare the advantages and disadvantages of MIR-DRIFTS and NIR in soil SOM analysis. If you are the head of a provincial soil survey project and need to complete SOM determination of 50,000 samples within 6 months, which method would you choose?

4. In a corn NIR spectrum, an abnormally strong absorption appears at 1940 nm (5155 cm⁻¹). What could be the cause? What impact will it have on the quantification of other components (protein, fat)? How to handle it?

5. Integrating the content of Ep 21–35, please explain why 'sampling method selection' is so critical in infrared industry applications? Give two examples from different industries.

References

[1] Janik L J, Skjemstad J O, Raven M D. "Characterization and Analysis of Soils Using Mid-Infrared Partial Least-Squares." Soil Biology & Biochemistry, 2008, 40(2): 412–423. DOI:10.1016/j.soilbio.2007.09.023.

[2] Williams P, Norris K. Near-Infrared Technology in the Agricultural and Food Industries. 3rd ed. AACC International Press, 2001. ISBN: 978-1-891127-37-6.

[3] Reeves J B. "Near- versus Mid-Infrared Diffuse Reflectance Spectroscopy for the Analysis of Soils." Applied Spectroscopy Reviews, 2010, 45(4): 289–322. DOI:10.1080/05704928.2010.486667.

[4] Demyan M S et al. "Soil Organic Matter Determination Using Mid-Infrared Spectroscopy." Soil Biology & Biochemistry, 2012, 52: 31–40. DOI:10.1016/j.soilbio.2012.04.008.

[5] Nocita M et al. "Soil Spectroscopy: An Alternative to Wet Chemistry for Soil Monitoring." Advances in Agronomy, 2015, 132: 139–158. DOI:10.1016/bs.agron.2015.06.002.

[6] Garcia J P, Sarmah A K, Bennett J McL. "ATR-FTIR Spectroscopy for Rapid Identification of Fertiliser Formulations and Detection of Adulteration." Talanta, 2019, 195: 535–542. DOI:10.1016/j.talanta.2018.11.072.

[7] Food and Agriculture Organization (FAO). Fertilizer Manual. 3rd ed. Kluwer Academic Publishers, 1998. ISBN: 978-0-7923-5032-7.

[8] Sanabria S et al. "Adulteration Detection in Fertilizers Sold in Sub-Saharan Africa Using Portable FTIR." Food Policy, 2021, 102: 102047. DOI:10.1016/j.foodpol.2021.102047.

[9] Williams P C. "Implementation of Near-Infrared Technology." In: Near-Infrared Technology in the Agricultural and Food Industries, 2001, Ch. 8: 145–169.

[10] FOSS. Infratec 1241 Grain Analyzer Reference Manual. 2020.
https://www.fossanalytics.com/

[11] China Instrument and Control Society. White Paper on the Application of Near-Infrared Spectroscopy in Agriculture. 2022.

[12] Mulla D J. "Twenty Five Years of Remote Sensing in Precision Agriculture: Key Advances and Remaining Knowledge Gaps." Biosystems Engineering, 2013, 114(4): 358–371. DOI:10.1016/j.biosystemseng.2012.08.009.

[13] Lobell D B, Field C B. "Global Scale Climate–Crop Yield Relationships and the Impacts of Recent Warming." Environmental Research Letters, 2007, 2(1): 014002. DOI:10.1088/1748-9326/2/1/014002.

[14] Cécillon L et al. "A Mid-Infrared Spectroscopic Database for Large-Scale Soil Surveys." Scientific Data, 2021, 8: 218. DOI:10.1038/s41597-021-01002-0.

[15] Mid-infrared soil organic matter spectral modeling and regional applications: Please search peer-reviewed literature such as Soil & Tillage Research / Geoderma; the specific DOI from the earlier draft could not be confirmed via Crossref, so unverifiable entries are not retained.

[16] ftir.fun Alkyl C-H functional group page. https://ftir.fun/ir/group/alk…

[17] ftir.fun Water molecule functional group page. https://ftir.fun/ir/group/wat…

[18] ftir.fun Siloxane functional group page. https://ftir.fun/ir/group/sil…

[19] ISO 11277:2020. Soil Quality — Determination of Particle Size Distribution in Mineral Soil Material.
https://www.iso.org/standard/…

Next Episode Preview: Ep 36 — Micro-FTIR: Seeing the Chemical Information of the Microscopic World
This concludes the intermediate industry applications! In the next episode, we enter Part 4 "Advanced — Cutting-Edge Technologies". Ep 36 starts with micro-FTIR (μ-FTIR): How does an infrared microscope combine with FTIR? What are the differences between transmission, reflection, and ATR micro-FTIR? How is spatial resolution limited by the diffraction limit (~10 μm @ 1000 cm⁻¹)? How does synchrotron radiation overcome this? What are the applications of MCT single-point vs. FPA array detectors?

This article is licensed under CC BY-NC-SA 4.0. Images are from public domain or labeled online sources, copyright belongs to the original authors.