Humidity Detector Based on Quartz Crystal Oscillator

Introduction

Background and Theoretical Inspiration

Accurate humidity monitoring is a fundamental requirement in numerous industrial, meteorological, and scientific applications. While traditional humidity sensors largely rely on capacitive or resistive mechanisms, they often exhibit limitations regarding response time, hysteresis, and long-term stability. To address the demand for high-precision, real-time sensing, mass-sensitive acoustic wave devices have garnered significant attention.

The theoretical foundation of our project is inspired by the Quartz Crystal Microbalance (QCM) principle. At its core, this principle relies on the mass-loading effect described by the Sauerbrey equation

$\mathrm{\Delta }f=-\frac{2f_0^2}{A\sqrt{{\rho }_q{\mu }_q}}\mathrm{\Delta }m$

Where a change in the mass ($\mathrm{\Delta }m$) attached to the surface of a piezoelectric quartz crystal results in a directly proportional, negative shift in its resonant frequency ($\mathrm{\Delta }f$). While commercial QCM systems utilize highly specialized and often expensive internal circuitry, the fundamental physical phenomenon—that a crystal's oscillation frequency drops as it gets heavier—serves as the conceptual basis for our humidity detector.

Sensing Mechanism: Water Glass Coating

To adapt a standard quartz crystal for humidity sensing, it must be functionalized with a hygroscopic material. In this project, Sodium Silicate ($Na_2SiO_3$), commonly referred to as water glass, is utilized as the sensitive coating. Water glass is a highly hydrophilic inorganic polymer that, when applied to the crystal surface and dried, forms a rigid, porous film rich in silanol (Si-OH) groups .

As the ambient relative humidity (RH) increases, water molecules are readily adsorbed onto the water glass layer via hydrogen bonding and capillary condensation. This accumulation of moisture effectively increases the mass loading on the quartz crystal. Consequently, following the Sauerbrey principle, the mechanical resonance frequency of the crystal decreases. This straightforward physical transduction mechanism allows us to map ambient humidity levels directly to measurable frequency shifts.

Core Architecture: The Colpitts Crystal Oscillator

The central engineering challenge of this project lies in designing a robust electronic interface capable of driving the quartz crystal and precisely tracking its frequency changes. Rather than relying on commercial QCM equipment, this project focuses on the implementation of a custom-built Colpitts crystal oscillator .

The Colpitts topology is highly regarded for its exceptional frequency stability and low phase noise, making it an ideal choice for sensor applications where minute frequency deviations must be detected. Our circuit utilizes a bipolar junction transistor (specifically the 2N2222 BJT) as the active gain element to sustain oscillation. A 6 MHz quartz crystal is integrated into the feedback loop. In this configuration, the crystal acts as a highly selective inductive element that dictates the oscillation frequency.

A critical aspect of the circuit design is the careful selection of feedback capacitors and biasing resistors to ensure reliable startup and to suppress any high-frequency parasitic oscillations that the 2N2222 might otherwise introduce due to its wide bandwidth. By ensuring the circuit oscillates cleanly at the crystal's fundamental 6 MHz mode, the system can reliably translate the mass variations from the water glass coating into a stable, measurable output signal.

Project Objectives and Scope

The primary objective of this project is to construct and characterize a functional, low-cost humidity detector driven by a custom Colpitts oscillator. The specific scope of work includes:

1. Design and hardware implementation of a 6 MHz Colpitts crystal oscillator using a 2N2222 transistor, ensuring stable and parasitic-free operation. 2. Application of a uniform sodium silicate (water glass) sensing layer onto the quartz crystal. 3. Calibration of the sensor by exposing it to environments with varying relative humidity and recording the corresponding frequency shifts. 4. Analysis of the sensor's performance metrics, including sensitivity, linearity, and response time.

Through this approach, the project aims to demonstrate that a fundamental electronic oscillator circuit, combined with a simple hygroscopic coating, can effectively replicate the core functionality of complex mass-sensitive humidity detectors.

Results and Discussion

Building Colpitts oscillator circuit

A Colpitts oscillator circuit was successfully constructed on a breadboard, and stable oscillation was achieved. The designed oscillation frequency was approximately 6 MHz. For the uncoated quartz crystal, the measured oscillation frequency was 5.9786 MHz, which is consistent with the expected value within experimental uncertainty.

Colpitts oscillator circuit on breadboard

Waveform Diagram without coating

Hygroscopic Film Coating

To enable humidity sensing, a hygroscopic film was deposited on the quartz crystal using a sodium silicate (water glass) solution. The original solution (density $1.39g/c{m}^3$) was first diluted to an appropriate concentration (origin solution : DI water = 1: 500). Subsequently, approximately 6 μL of the diluted solution was drop-cast onto the center region of the quartz crystal, while the gold electrodes at the periphery were intentionally left uncovered to avoid interference with electrical conduction. The coated crystal was then placed on a hot plate and heated gradually to 60–80 °C, where it was maintained for approximately one hour to evaporate the solvent and form a solid hygroscopic thin film.

Dropped solution on the quartz crystal

Heating on the hot plate

Sustainability Test

After film deposition, the stability of the oscillator was evaluated under different environmental conditions. Measurements were conducted in a laboratory environment (22 °C, ~40% relative humidity) and an outdoor environment (30 °C, ~75% relative humidity). In both cases, the oscillator was operated continuously for approximately one hour, during which stable oscillation with negligible frequency drift was observed, indicating good operational stability.

Outdoor environment testing

Humidity Sensing Performance

The humidity sensing performance was then investigated. Due to the lack of a calibrated humidity chamber, a qualitative humidity variation was introduced by placing different volumes of deionized (DI) water inside a sealed test tube then place the crystal on the top of it. The oscillator frequency was monitored under these conditions. Given that the relative frequency shift is small compared to the absolute frequency (~6 MHz), the frequency was estimated by recording the waveform over a 1-second interval and counting the number of oscillation cycles. The results show a clear dependence of oscillation frequency on the amount of water present. Starting from 10 mL of DI water, The oscillate frequency of the origin crystal in lab environment is 5.981744 MHz, which is the same as when placed at the top of the sealed test tube with 10 mL of DI water. Every additional 10 mL led to an approximate frequency decrease of 25 Hz, when in 20mL, the frequency is 5.981719 MHz; when in 30mL, the frequency is 5.981694 MHz. This trend indicates that the system is sensitive to humidity-induced changes.

Frequency-Volume Curve

Conclusion and Outlook

This work demonstrates the feasibility of using a quartz crystal oscillator as a humidity sensor. The observed frequency shift can be attributed to the mass-loading effect on the quartz crystal: as water is absorbed by the hygroscopic water-glass film, the effective mass on the crystal surface increases, resulting in a decrease in resonance frequency. This sensing mechanism is consistent with the Sauerbrey relation commonly used to describe quartz crystal microbalance behavior. Overall, the results show that a water-glass-coated quartz crystal oscillator is capable of responding to humidity changes. However, the performance of the system is strongly influenced by the coating condition, measurement temperature, and the very small magnitude of the frequency shift relative to the 6 MHz baseline. One important factor is the thickness of the water-glass coating. In principle, a thicker sodium silicate film can absorb more water from the environment and therefore produce a larger mass-loading effect, leading to a greater frequency shift. However, this advantage is accompanied by a limitation: as the film becomes thicker, the additional mass and mechanical damping imposed on the quartz crystal also increase. If the coating is too thick, the oscillation amplitude may decrease and the oscillator may become more difficult to sustain. In the present work, the sensing layer was prepared by drop-casting diluted water glass onto the crystal surface. Although this method is simple and practical, it does not guarantee a perfectly uniform coating thickness. As a result, local nonuniformity of the film may contribute to measurement variation. Therefore, the effect of coating thickness can be understood as a trade-off: a film that is too thin may produce only a weak humidity response, while a film that is too thick may degrade oscillator performance. Optimizing the thickness and uniformity of the sensing layer is thus important for improving both sensitivity and reliability. Temperature is another critical factor in this experiment. Under normal testing conditions, the oscillator response was generally stable, indicating that the circuit and coated crystal were capable of reliable operation for qualitative sensing. However, a clear difference was observed between cold-water and hot-water testing. When cold water was used, the frequency change remained small but stable. In contrast, when hot water at around 60 °C was used, the oscillation could no longer be measured reliably, and the waveform became nearly flat. This behavior suggests that the hot-water condition did not simply enhance the humidity response, but instead pushed the sensor beyond its stable operating range. A possible explanation is localized vapor condensation near the quartz crystal or surrounding circuit. Under such conditions, the sensor surface may no longer behave as a lightly mass-loaded rigid film, but rather as a liquid-coupled or highly viscoelastic layer. This would strongly increase damping of the quartz oscillation, reduce the resonance quality factor, and prevent reliable frequency detection. In addition, hot-water testing introduces coupled effects of humidity and temperature drift, making the result unsuitable for direct quantitative interpretation. Therefore, temperature should be regarded as a significant disturbance factor in the present setup, especially under extreme vapor conditions. A further limitation of the system is the very small magnitude of the sensing signal. The quartz crystal oscillates at approximately 6 MHz, whereas the humidity-induced frequency shift is only on the order of tens to around one hundred hertz. Although this shift is meaningful and consistent with the expected mass-loading effect, it represents only a very small fractional change compared with the carrier frequency. Consequently, the measurement is highly sensitive to instrumental resolution, baseline drift, temperature fluctuation, and waveform noise. This means that the difficulty lies not in the absence of a sensing response, but in the challenge of accurately detecting a very small change superimposed on a much larger base frequency. For this reason, a high-resolution frequency counter or a more stable digital frequency measurement system would be more appropriate than relying mainly on simple oscilloscope observation.