Scientific Exploration at Pitt-Bradford
An independent, internship-like scientific experience involving SEM/EDS analysis, an Arduino and LiDAR build, and evidence-led investigation, made possible by Dr. Ovidiu Franţescu.
Abstract
From August 24-29, 2026, with the main project work completed through August 28, I took part in an independent, internship-like scientific experience at the University of Pittsburgh at Bradford. I learned a scanning electron microscope workflow, used EDS results to investigate a geode, built an Arduino and LiDAR warning system, attended science classes, and practiced approaching unfamiliar problems through evidence and iteration.
Method
Dr. Ovidiu Franţescu made all of the SEM work possible by providing instrument access, teaching the workflow, explaining the underlying science, and allowing me to investigate selected regions. Across two sessions I helped prepare and mount samples, navigated specimens, adjusted magnification and focus, captured 28 retained images, made measurements, and completed 10 retained SEM/EDS geode analyses under his guidance.
Findings
The clearest data-supported finding was that small geode regions could have very different reported elemental compositions. Retained analyses included regions led by silicon, oxygen, tin, or antimony. The more important result was methodological: visual irregularities became questions that could be tested with local elemental evidence.
Limitations
The SEM work consisted of two sessions of roughly one to two hours each. EDS measurements represent selected spots, lines, or mapped regions rather than the bulk composition of the geode. Some observations remained exploratory, and biological sample coating was performed by Dr. Franţescu rather than independently.
Next questions
A future session would use repeated selection criteria, record particle or crystal size alongside each EDS result, and test whether the apparent relationship between structure and composition is repeatable. I would also continue learning advanced SEM functions and the detector physics behind image formation.
Experience at a glance
- Experience dates
- Aug. 24-29, 2026
- SEM sessions
- 2
- Retained SEM images
- 28
- EDS projects
- 10
- Engineering builds
- 1
Main project work continued through August 28.
Roughly one to two hours each.
11 from the first session and 17 from the second.
Spot, map, and line-scan work retained from the geode investigation.
An Arduino, LiDAR sensor, LED, and a CAD-designed enclosure.
Faculty contribution
- SEM mentor and faculty host
- Dr. Ovidiu Franţescu
Every SEM image and EDS result in this report was possible because Dr. Franţescu provided access, instruction, scientific context, and the freedom to investigate selected regions.
Biological structures from the first session
From visual irregularity to elemental evidence
Data
Largest reported signal in selected EDS results
The largest reported local weight-percentage signal varied between selected geode regions. These values are semiquantitative and do not represent the composition of the whole specimen.
| Label | Value |
|---|---|
| Geode 2 - Si | 40.1 % |
| Geode 4 - O | 23.5 % |
| Geode 5 - Sn | 56.3 % |
| Geode 8 - Sb | 41.5 % |
| Geode 10 - Sb | 35.8 % |
Challenge
What needed to change
Move beyond simply recording interesting SEM images and decide which small surface differences were worth testing.
Solution
How I approached it
Use cracks, boundaries, divots, distinct crystals, and apparent changes in depth to choose EDS locations, then compare the elemental result with the original visual expectation.
Continue exploring
Read the complete report and faculty profile
The nine-page report includes the complete reflection, retained data inventory, limitations, and full EDS summary.