high school physics experiments

6 Fun Physics Experiments For High School Students

Physics as a science is not just about nuclear fission or fusion, or just unraveling the mysteries of the Big Bang theory! It is in things as simple as the apple falling from a tree or a school bus in motion (or even in a state of inertia or rest). 

It is thus important to teach the high schoolers, physics in their everyday routine. Also exciting is to design and demonstrate physics experiments from their everyday items.

After all, the foundational understanding of physics in their young minds should not be just a textbook view, but something they can perform on their own. 

Of course, we are here to help with a list of physics experiments you can perform with high schoolers to open their minds to the new possibilities that abound in the universe.

Importance of conducting physics experiments for high schoolers

Physics experiments are not just another activity on your school core card. Their significance goes way beyond it and is crucial for high school students for several reasons:

• Developing problem-solving skills:   Experiments allow students to apply their physics knowledge and develop their problem-solving abilities. They can better understand the concepts by seeing the results firsthand.
• Enhancing critical thinking:  Doing experiments encourages students to think critically, analyze data, and draw conclusions. This helps them solve complex problems and make better decisions.
• Connecting theory and practice:  Physics experiments enable students to apply their theoretical knowledge in practice and make connections between theory and reality, improving their understanding of physics concepts.
• Stimulating creativity:  Experiments in the lab can stimulate creativity as students are encouraged to think outside the box when trying to solve problems. This encourages innovation and unique solutions.
• Increasing confidence and interest: Experiments can help boost student confidence as they see the results of their efforts. This will also motivate them to keep exploring and developing their skills.
• Developing observational and practical skills:  Experiments require students to observe and analyze data and gain hands-on experience, which helps them develop important skills that will be useful throughout their lives.

Thus, these experiments go a long way in developing critical-thinking individuals who can connect theory with practice and change the world with a book in their hands!

Fun & Challenging Physics Experiments for High Schoolers

1. put together a mini tesla coil.

A Tesla coil is an electrical transformer circuit to produce high voltage and high frequency alternating current electricity.

But what if we tell you that you can make your very own mini Tesla coil in your school lab or house backyard? 

No, you don’t need expensive and hard-to-get materials to get started.  Just a little physics knowledge and being handy with old stuff and you get your very own Tesla coil!

 Material required to build the mini Tesla coil:

• A high-voltage transformer or power supply. You can even use a flyback transformer from an old television set. This provides the high voltage input.
• A primary coil is typically made of a few turns of thick copper wire or tubing. This is connected to the high-voltage input.
• A secondary coil is made of hundreds of turns of thin enameled wire wound around a non-conductive form like a PVC pipe. This creates a high-voltage output.
• A capacitor bank creates the resonant circuit with the secondary coil.
• A spark gap to create the high-voltage discharge.

Once you have the material in handy, the next step is to assemble them together :

• Wind the secondary coil with the thin wire, ensuring the turns are evenly spaced.
• Connect the primary coil to the high-voltage input.
• Add the capacitor bank and spark gap to complete the resonant circuit.
• Carefully tune the circuit to achieve maximum voltage output.

Working with a high school student, it is important to start with a small-scale design and low power levels. You also have to take necessary precautions like proper insulation and grounding.

The Mini Tesla coil is ready with minimum materials and effort!

2. Remove the air in a DIY Vacuum chamber

Create your very own vacuum chamber with simple, everyday stuff. Let’s put those pots and lids in the service of science!

Materials required for DIY vacuum chamber

• Large cooking pot or container, 
• A piece of polycarbonate or acrylic sheet for the lid, 
• Silicone gasket maker, and 
• Various fittings like a vacuum gauge, ball valves, and bulkhead connectors.

The next steps to start assembling:

• Drill the necessary holes in the pot and lid to attach the fittings.
• Apply a silicone gasket maker around the rim of the pot to create an airtight seal with the lid.
• Connect the vacuum gauge, ball valves, and tubing to allow you to pull a vacuum and release it.
• Once assembled, you can use the vacuum chamber to degas silicone and remove air bubbles from resin castings.

Of course, a much more fun experiment would be to boil the water at room temperature due to reduced pressure in the chamber or even expand a marshmallow. The vacuum chamber opens endless possibilities for experiments!

3. Demonstrating the power of friction with sticky notes

The beautiful and multi-colored stacks of sticky notes are a fun addition to the school stationery. But guess what, they can be instrumental in understanding the power of friction as well!

Materials required for understanding friction:

• Pairs of sticky notes.
• Small everyday objects to use as weights.

Steps to the experiment:

• Interleave the pages of two stacks of sticky notes, say 40 interleaved pages.
• Put small, daily objects as weights on them to see how much they can support.

You will be surprised with the results! A stack of 40 interleaved notes can support up to 60 pounds! This demonstrates how friction can create a very strong “grip” between surfaces, even if the materials (like paper) have a relatively low coefficient of friction on their own.

It’s like a phonebook being used to lift a car!

4. Measure the speed of light using a microwave oven

Have a bar of chocolate with you? And some spare time in the kitchen with the microwave?

That’s all you need to calculate the speed of light!

Materials required for calculating the speed of light:

• Chocolate bar or some marshmallows.
• A microwave

Let’s start with the steps:

• Place the food item in the microwave, and spread it out from side to side. Cook it for 5-10 seconds, being careful not to burn it.
• Measure the distance between the melted spots on the food – this should be around 6 cm. Each spot is half a wavelength, so the full wavelength is 12 cm or 0.12 m.
• Look on the back of your microwave for the frequency, which is typically 2450 MHz.
• Use the formula: Speed of light = Frequency x Wavelength. Plugging in the values, you get: Speed of light = 2,450,000,000 Hz x 0.12 m = 294,000,000 m/s.

Now before you get all science-y on us, we know the speed of light is 300,000,000m/s.  But getting so close to the real value with an old microwave and melted chocolate is an experiment worth performing! And smugly announcing the results to the world!

5. Build a brighter light bulb

Want to impress Mom and Dad by building a homemade bulb brighter than the routine one? We have your back.

Materials required to build a brighter bulb:

• A thin filament
• Inert gas like C02
• Higher voltage power source, like 12-24V from multiple batteries in series

Let’s assemble our bulb:

• Use a thinner filament material like a single strand of iron wire or pencil lead (graphite) instead of a thick braided filament. The thinner filament will heat up more efficiently and glow brighter.
• Enclose the filament in a glass jar filled with an inert gas like CO2 to prevent the filament from burning out quickly. Blow CO2 into the jar using an air nozzle or place a piece of dry ice inside and let it sublimate. 
• Connect the filament to a higher voltage power source, like 12-24V from multiple batteries in series, to provide more current and heat the filament to a brighter glow. 

Although the bulb will not be as bright as a commercial LED, it will still be impressively brighter to increase the chances of a major cash flow into your college funds by your parents!

6. Boil water in a paper cup

Yes, you may seem surprised, but it’s very much possible!

Materials required to boil water in a paper cup:

• A heat source like a lamp or burner.

It’s as simple as it gets!

• Put water in a paper cup and place it on a burner.
• The water will indeed boil!
• This is possible because water has a high specific heat capacity, which means it absorbs heat from the cup faster than the paper can reach its ignition temperature. As the water heats up, it conducts heat away from the paper, preventing it from burning.
• The water also continues to absorb heat through convection until it reaches its boiling point of 100°C, at which point the temperature of the water remains relatively constant.

However, the key word to remember here is ‘paper’, and not those styrofoam cups! Styrofoam is an insulator and does not conduct heat well, so the Styrofoam cup will disintegrate before the water boils.

However, to give you another piece of our physics secret, to boil water in a Styrofoam cup, you can use a hot rock placed in the water to transfer heat more effectively!  Start experimenting, thank us later!

Once your high school students are done with these fun experiments, they will be better able to appreciate the principles of physics in their everyday lives. And you may never know, the kid you thought slow, who always sits on the back bench of the class brooding, might just be another Einstein in the making!

An Engineer, Maths expert, Online Tutor, and animal rights activist. I have more than 5 years of teaching experience and have worked closely with students with learning disorders. I have worked with special educators, counselors, and experts in dealing with common issues that students face during their academic journey.

Leave a Comment Cancel reply

Save my name, email, and website in this browser for the next time I comment.

30 Best Science Experiments & Projects for High School

Welcome to our round-up of top science fair projects and science experiments tailored specifically for curious high school students.

Science fair is not just about the glitz and glamour of a first-place trophy; it’s about the passion, the inquiry, and the insatiable curiosity that drive every scientist, young and old. Hopefully, our curated list of the best hands-on science fair projects for high school students will ignite that curiosity in you.

Each project on this list offers a unique opportunity to dive deep into scientific inquiry and present findings with both clarity and flair.

Let’s dive in and make learning an unforgettable adventure!

1. Burn Calories

Don’t miss this opportunity to unravel the mysteries of energy transformation and uncover the scientific secrets hidden in the simplest of substances!

Learn more: Science Buddies

2. Extracting DNA from Strawberry

By following a series of simple yet insightful steps, students will witness the magical moment of DNA extraction, fostering a deeper appreciation for the fundamental building blocks of life.

Learn more: Extracting DNA from Strawberry

3. Build a Simple DIY Newton’s Cradle

As students assemble the materials and witness the rhythmic dance of swinging spheres, they will witness the scientific principles they’ve learned in the classroom come to life before their eyes.

4. Make a Monster Dry Ice Bubbles

Unleash your inner mad scientist and learn how to make Monster Dry Ice Bubbles with this high school science experiment!

Get ready to be captivated as you create giant, spooky bubbles that dance and swirl with the mysterious power of dry ice.

Learn more: Wonder How To

5. Soil Erosion Experiment

As stewards of our environment, it’s crucial to comprehend the impact of natural processes like soil erosion.

Through this experiment, students will gain a deeper appreciation for the significance of soil conservation and sustainable land management practices.

Learn more: Life is a Garden

6. Candle Carousel

This experiment combines the wonders of physics with the art of crafting, making it an enriching experience that ignites curiosity and fosters a deeper appreciation for the elegant dance of energy in our world.

7. Find Out if Water Conducts Electricity

In this captivating activity, students will explore the conductive properties of water and unlock the secrets of how electrical currents flow through different substances.

Learn more: Rookie Parenting

8. Roller Coaster Stem Experiment

By experimenting with various designs and track configurations, students will refine their problem-solving skills and gain valuable insights into the practical applications of physics and engineering.

Learn more: STEM Project

9. Lemon Battery

Engaging in this experiment not only teaches the basics of electrical circuits but also sparks curiosity about the natural world and the science behind it.

Learn more: Coffee Cups and Crayons

10. Watering Plants Using Different Liquids

Discover the wonders of plant hydration with the intriguing high school science experiment – “Watering Plants Using Different Liquids.” In this captivating project, students explore how various liquids impact plant growth and health.

Learn more: Lemon Lime Adventures

11. Measure Electrolytes Found in Sports Drinks

By conducting a series of tests and analyses, students will quantify the electrolyte content present in various sports drinks.

12. Relight the Flame Without Directly Touching It

This captivating project challenges students to learn about the intriguing properties of heat transfer and combustion.

By exploring different methods to reignite a candle flame without physical contact, students will uncover the secrets of heat conduction, convection, and radiation.

Learn more: Stevespangler

13. Conduct Fingerprint Analysis

This captivating project immerses students in the intriguing world of crime scene investigations, where they will uncover the uniqueness of fingerprints and their role in forensic science.

14. Separate Water Into Hydrogen And Oxygen Using Electrolysis

This electrifying project allows students to explore electrolysis and the decomposition of water into its elemental components.

Learn more: Navigating by Joy

15. Simple Color Detection Circuit 

This experiment not only introduces fundamental concepts in electronics and circuitry but also opens up endless possibilities for real-life applications, from automated sorting systems to color-sensitive devices.

16. Carbon Sugar Snake

This enchanting project allows students to witness a dazzling display of science as they combine common household ingredients to create a dark, coiling “snake” made of carbon.

Learn more: Kiwi Co

17. Build a Hydraulic Elevator

This captivating project invites students to learn about engineering and fluid mechanics. By constructing a working model of a hydraulic elevator, students will explore the principles of Pascal’s law and the fascinating concept of fluid pressure.

Learn more: Teach Beside Me

18. Brew up Some Root Beer

This enticing project invites students to explore the fascinating world of chemistry and fermentation while creating their own delicious and bubbly concoction.

Learn more: Home School Creations

19. Extracting Bismuth From Pepto-Bismol Tablets

This hands-on experiment not only sheds light on the principles of chemistry and lab techniques but also highlights the real-world applications of bismuth in medicine and various industries.

Learn more: Popscie

20. Solar-Powered Water Desalination

By designing and building a solar-powered water desalination system, students will learn how to harness the sun’s energy to purify saltwater and make it safe for consumption.

21. Applying Hooke’s Law: Make Your Own Spring Scale

By designing and constructing their very own spring scale, students will uncover the principles of Hooke’s Law and the relationship between force and displacement in a spring system.

22. Homemade Hand Warmer

By creating their own hand warmers using safe and easily accessible materials, students will witness the magic of heat generation through chemical processes.

Learn more: Steve Spangler

23. Explore the Concept of Symbiosis Involving Nitrogen-Fixing Bacteria.

By investigating how certain plants form a mutually beneficial bond with these bacteria, students will gain insights into the essential role of nitrogen fixation in the ecosystem.

Learn more: Education.com

24. Center of Gravity Experiment

This fascinating project invites students to explore the concept of the center of gravity and its role in determining stability.

25. Power up Homemade Batteries

This captivating project invites students to learn about electrochemistry and energy generation.

Learn more: 123 Homeschool

26. Film Canister Explosions

Prepare for a blast of excitement and chemistry with the high school science experiment – “Film Canister Explosions!” This project teaches students about chemical reactions and pressure build-up.

27. Investigating Osmosis with Potato Slices

This hands-on experiment not only provides a practical understanding of osmosis but also highlights its relevance in everyday life, from understanding plant hydration to food preservation techniques.

28. Make Homemade Fly Trap

This captivating “Make Homemade Fly Trap!” project invites students to explore the principles of pest control and observe the behavior of flies.

29. Hydroponics: Gardening Without Soil

This exciting project invites students to explore innovative agricultural practices that harness water and nutrient solutions to grow plants.

By setting up their hydroponic system and nurturing plants through this method, students will witness the fascinating dynamics of root development and nutrient absorption.

30. Clothespin Airplane

As they test and modify their creations, students will learn about the principles of lift, thrust, and drag, gaining a deeper understanding of how these forces come together to keep airplanes soaring through the skies.

Learn more: Steamsational

Similar Posts:

• 68 Best Chemistry Experiments: Learn About Chemical Reactions
• 37 Water Science Experiments: Fun & Easy
• Top 40 Fun LEGO Science Experiments

Leave a Comment Cancel reply

Save my name and email in this browser for the next time I comment.

• Sign in / Register
• Administration
• Edit profile

The PhET website does not support your browser. We recommend using the latest version of Chrome, Firefox, Safari, or Edge.

Sciencing_Icons_Science SCIENCE

Sciencing_icons_biology biology, sciencing_icons_cells cells, sciencing_icons_molecular molecular, sciencing_icons_microorganisms microorganisms, sciencing_icons_genetics genetics, sciencing_icons_human body human body, sciencing_icons_ecology ecology, sciencing_icons_chemistry chemistry, sciencing_icons_atomic & molecular structure atomic & molecular structure, sciencing_icons_bonds bonds, sciencing_icons_reactions reactions, sciencing_icons_stoichiometry stoichiometry, sciencing_icons_solutions solutions, sciencing_icons_acids & bases acids & bases, sciencing_icons_thermodynamics thermodynamics, sciencing_icons_organic chemistry organic chemistry, sciencing_icons_physics physics, sciencing_icons_fundamentals-physics fundamentals, sciencing_icons_electronics electronics, sciencing_icons_waves waves, sciencing_icons_energy energy, sciencing_icons_fluid fluid, sciencing_icons_astronomy astronomy, sciencing_icons_geology geology, sciencing_icons_fundamentals-geology fundamentals, sciencing_icons_minerals & rocks minerals & rocks, sciencing_icons_earth scructure earth structure, sciencing_icons_fossils fossils, sciencing_icons_natural disasters natural disasters, sciencing_icons_nature nature, sciencing_icons_ecosystems ecosystems, sciencing_icons_environment environment, sciencing_icons_insects insects, sciencing_icons_plants & mushrooms plants & mushrooms, sciencing_icons_animals animals, sciencing_icons_math math, sciencing_icons_arithmetic arithmetic, sciencing_icons_addition & subtraction addition & subtraction, sciencing_icons_multiplication & division multiplication & division, sciencing_icons_decimals decimals, sciencing_icons_fractions fractions, sciencing_icons_conversions conversions, sciencing_icons_algebra algebra, sciencing_icons_working with units working with units, sciencing_icons_equations & expressions equations & expressions, sciencing_icons_ratios & proportions ratios & proportions, sciencing_icons_inequalities inequalities, sciencing_icons_exponents & logarithms exponents & logarithms, sciencing_icons_factorization factorization, sciencing_icons_functions functions, sciencing_icons_linear equations linear equations, sciencing_icons_graphs graphs, sciencing_icons_quadratics quadratics, sciencing_icons_polynomials polynomials, sciencing_icons_geometry geometry, sciencing_icons_fundamentals-geometry fundamentals, sciencing_icons_cartesian cartesian, sciencing_icons_circles circles, sciencing_icons_solids solids, sciencing_icons_trigonometry trigonometry, sciencing_icons_probability-statistics probability & statistics, sciencing_icons_mean-median-mode mean/median/mode, sciencing_icons_independent-dependent variables independent/dependent variables, sciencing_icons_deviation deviation, sciencing_icons_correlation correlation, sciencing_icons_sampling sampling, sciencing_icons_distributions distributions, sciencing_icons_probability probability, sciencing_icons_calculus calculus, sciencing_icons_differentiation-integration differentiation/integration, sciencing_icons_application application, sciencing_icons_projects projects, sciencing_icons_news news.

• Share Tweet Email Print
• Home ⋅
• Science Fair Project Ideas for Kids, Middle & High School Students ⋅

Easy High School Physics Experiments

Light-Dispersion Experiments for Kids

Physics may seem like an intimidating subject, but there are ways to make it fun. Not only is it the foundation for other sciences like chemistry and meteorology, it also explains so much about the world we live in. Physics explores the fundamental concepts of matter, energy, space and time, and the interactions between these properties. For high school students looking for simple experiments, light, static electricity and thermodynamics are great places to start.

TL;DR (Too Long; Didn't Read)

Shine a flashlight through water and milk to discover why the sky is blue but the sunset is red; use a comb to bend water with static electricity; and watch a hard-boiled egg get sucked into a bottle to see thermodynamics in action.

The Color of Light

Ever wonder why the sky is blue, but the sunset is red? Use a flashlight, a transparent rectangular container, water and a cup of milk to find out why.

Fill the container three-quarters full of water and shine the flashlight into the side of the container. Observe the light from the opposite side and the end of the container. At most, a few white dust particles might be seen where the beam passes through.

Now stir 1/4 cup of milk into the water. Observe the light from the opposite side and the end of the container. From the other side, the light may seem blue, and from the end, the light may seem yellow. Note the width of the beam. Repeat until all of the milk is added. You'll notice after each addition that the blue darkens, the yellow turns to orange and the width of the beam increases.

So, why does the light appear two different colors, depending on the angle? Light travels in a straight line unless it encounters particulates that cause the beam to scatter. The more milk (which contains fat and protein particles) you add into the water, the more the light scatters, with the blue bending while the red and orange continue in more of a straight line. As for the sunset, because of the sun's path, the light has farther to travel at that time and encounters more dust particles in the atmosphere.

Static Electricity

Static electricity can shock an unsuspecting person, and it can also move objects. Use a nylon comb and a faucet to watch static electricity bend water.

Turn the faucet on so that water 1/16 inch in diameter flows from the tap. Run the comb through hair a few times. Hold the comb 3 to 4 inches below the tap with the teeth of the comb an inch from the water stream. Note what happens. Move the comb closer and observe what happens. Run the comb through hair again and see if it changes the results. Try adjusting the water stream to see if it makes a difference. Finally, try different sized combs and repeat.

Combing hair creates static electricity. One object becomes negatively charged by gaining electrons, while the other object becomes positively charged by losing electrons. The stream of water moves toward the comb because electrons from the water are attracted to the charged comb. The combed hairs might also repel one another, since each strand holds the same charge, and like charges repel.

High and Low Pressures

What does the weatherman mean by "high pressure" and "low pressure"? A hard-boiled egg, an old-fashioned glass milk bottle and some matches can help you find out.

Peel a cooled, hard-boiled egg. Simultaneously light three matches and drop them into an empty glass bottle. Quickly cover the opening with the egg. After the matches extinguish, watch the egg get sucked into the bottle.

The heat from the matches causes the air sealed in the bottle to expand. After the matches go out, the air cools and contracts. The pressure inside the bottle becomes lower than the pressure outside the bottle. As the pressure equalizes, the egg squeezes into the bottle.

Fascinating stuff! Enjoy these experiments, and hopefully these physics concepts become a little easier to digest.

Related Articles

Light wave experiments, how does light travel from the sun to earth, experiments with a magnifying glass, how to make rainbows with prisms, how to diffuse a laser beam, how to create a prism, which colors reflect more light, experiments with heat radiation, how to make glowing water without a black light, how to make a homemade black light, what colors attract heat, easy indoor rainbow experiments, food coloring experiments, science projects about rainbows, how to use a vivitar telescope, easy 10 minute science projects, science fair projects with fiber optics, how to make glowing water for a science fair project.

About the Author

Michele Jensen started writing professionally for businesses in 1999. Her writings include articles for eHow, Answerbag and COD, marketing materials and project-related documentation. She received her Bachelor of Science degree in electrical engineering from the University of Houston and a Master of Science degree in international relations from Troy State University.

Find Your Next Great Science Fair Project! GO

• Earth Science
• Physics & Engineering
• Science Kits
• Microscopes
• Science Curriculum and Kits
• About Home Science Tools

Resource Center Home > Science Projects > Science Fair Projects > Physics Science Fair Projects

Physics Science Fair

Find physics science fair project ideas about magnetism, electricity, energy and solar power, and more.

Do this spinoff of the elephant toothpaste experiment using household items like yeast and hydrogen peroxide.

Split water into hydrogen and oxygen gas using two pencils and a battery in this fun electrolysis science project!

Learn about physics as you build your own mousetrap marshmallow catapult with this science project.

Learn about electromagnets and magnetic levitation.

A brief guide to exceptional science projects and science project videos on the web.

Build a mini solar car to see how to use solar energy for power.

Make a balloon rocket car and watch a video showing the project in action.

Return to PocketLab Home Page

High School Physics

Prepare your students for real-world problem solving and open-ended lab experiments. Experienced educators and curriculum specialists have developed each of these lessons, and we have tested them in real classrooms. PocketLab physics lessons cover introductory and advanced topics from one-dimensional motion to electricity and magnetism to simple harmonic motion. Browse all the high school and AP-level physics lessons below or use the filters to search for specific content.

Rolling Resistance Physics Lab

Rolling Resistance

Rolling resistance is a force that opposes the motion when an object rolls along a surface.There are many examples of objects experiencing rolling resistance: car or bicycle tires on pavement, skateboard wheels on a half pipe ramp, steel wheels on a railroad track, ball bearings in a pulley, bowling balls on a bowling lane, and carts rolling on a dynamics track, just to mention a few. Many factors can affect the magnitude of the forces associated with rolling resistance.

Moment of Inertia Challenge

The Moment of Inertia Challenge

Determinig the moment of inertia about the center of mass mathematically  of an object with complex geometry is not an easy task.  Consider, for example, the empty 3D filament reel shown in Figure 1, an empty 1-kg Polymaker Polylite ™ PLA reel.  With its holes and intricate axle design, the best way to determine the moment-of-inertia about its center of mass is via a laboratory experiment.  Voyager (or PocketLab One) is mounted to the reel for data collection.  

Kinematics of Translational and Rotational Motion

Introduction

Empty 3D filament reels are great devices to use in the physics classroom.  There's a good chance that you and your students could come up with some interesting physics lab investigations using these reels.  Attach Voyager or PocketLab One to the reel as shown in Figure 1, and the possibilities are endless!  This lesson describes a lab in which your students study the kinematics of both translational and rotational motion when the reel rolls down a ramp on its axle.  Students are often surprised when they see the reel speed up upon reaching the floor on which

Moment of Inertia of a 3D Filament Reel About Its CM

Your school can put all of those empty 3D filament reels to good use in the physics classroom.  There's a good chance that you and your students could come up with some interesting physics lab investigations using these reels.  As shown in Figure 1, attach Voyager or PocketLab One to the reel, and the possibilities are endless!  This lesson describes a lab in which your students determine the moment of inertia of an empty 3D filament reel about its center-of-mass.  They will accomplish this using two independent methods.  One method has the reel acting as a p

Unrolling Spool Problem Quantitative Experiment

Think twice before discarding your school's empty 3D filament reels.  There's a good chance that you and your students could come up with some interesting physics lab investigations using these reels.  As shown in Figure 1, attach Voyager or PocketLab One to the reel, and the possibilities are endless!  This lesson describes a quantititive experiment that your students can perform in a study of the classic "unrolling spool problem".

Periodic Motion of an Empty 3D Filament Reel

Don't discard your school's empty 3D filament reels.  There's a good chance that you and your students could come up with some interesting physics lab investigations using these reels.  Attach Voyager or PocketLab One to the reel and the possibilities are endless!  This lesson describes a unique experiment in which periodic motion is investigated using an empty 3D filament reel.  Depending on the teacher's goals and amount of detail in the analysis of collected data, this lab could be used from the 4th grade through high school.  The

NGSS Seismic Basketball Challenge

The NGSS Seismic Basketball Challenge

This NGSS seismic basketball challenge fits well in the study of motion for high school physics students.  Here is a statement describing the challenge:

Place PocketLab Voyager on a wood floor with accelerometer data being captured. Drop a basketball onto the floor near Voyager and let it bounce several times, being careful to not let it hit Voyager.  From the recorded accelerometer data, determine the original height from which the basketball was released.  

Intelino / Voyager Lab: "Floor-it" Acceleration/Max Speed

The purpose of this lesson is to challenge your students to design an experiment for which data from PocketLab Voyager is used to determine the "floor-it" acceleration and maximum speed of the intelino smart train engine.  Required data should be obtained in a single run of data collection by the PocketLab app.  Figure 1 shows a picture of Voyager attached to the top of an intelino smart engine.  Designed for all ages, intelino is intuitive with its app, has built-in sensors to provide an interactive experience for the user, and is easily programmed with colo

intelino / Voyager Lab: Stopping Distance vs. Speed

Have you ever been told not to follow too close to the driver ahead of you?  To keep a safe distance?  To abide by the "3-second rule"?  To keep a distance of at least one car length for every ten miles per hour of speed?  These questions all deal with the issue of stopping distance versus speed in order to avoid crashes.  A great way to investigate the relationship between stopping distance and speed is to interface Voyager with an " intelino® smart train ".   Designed for all ages, intelino is intuitive with its app, has bui

intelino / PocketLab: Velocity vs. Impulse to Stop

While driving at 40 mph, you see a red stop light ahead.  You press your brakes for several seconds, gradually coming to a stop.  A little later on the same road at 40 mph, you approach another light, this time green.  While approaching this light, it suddenly changes to yellow.  You make a split-second decision to put on your brakes to avoid going through a red light.  With the brakes applied quite hard, you quickly stop, waking up your sleeping friend in the front passenger seat.

To access this free lesson, please sign up to receive communications from us:

30 Best Science Experiments & Projects for High School

Welcome to our round-up of top science fair projects and science experiments tailored specifically for curious high school students.

Science fair is not just about the glitz and glamour of a first-place trophy; it’s about the passion, the inquiry, and the insatiable curiosity that drive every scientist, young and old. Hopefully, our curated list of the best hands-on science fair projects for high school students will ignite that curiosity in you.

Each project on this list offers a unique opportunity to dive deep into scientific inquiry and present findings with both clarity and flair.

Let’s dive in and make learning an unforgettable adventure!

1. Burn Calories

Don’t miss this opportunity to unravel the mysteries of energy transformation and uncover the scientific secrets hidden in the simplest of substances!

Learn more: Science Buddies

2. Extracting DNA from Strawberry

By following a series of simple yet insightful steps, students will witness the magical moment of DNA extraction, fostering a deeper appreciation for the fundamental building blocks of life.

Learn more: Extracting DNA from Strawberry

3. Build a Simple DIY Newton’s Cradle

As students assemble the materials and witness the rhythmic dance of swinging spheres, they will witness the scientific principles they’ve learned in the classroom come to life before their eyes.

4. Make a Monster Dry Ice Bubbles

Unleash your inner mad scientist and dive into a world of enchanting and eerie fun with this high school science experiment: Make Monster Dry Ice Bubbles!

Get ready to be captivated as you create giant, spooky bubbles that dance and swirl with the mysterious power of dry ice.

Learn more: Wonder How To

5. Soil Erosion Experiment

As stewards of our environment, it’s crucial to comprehend the impact of natural processes like soil erosion.

Through this experiment, students will gain a deeper appreciation for the significance of soil conservation and sustainable land management practices.

Learn more: Life is a Garden

6. Candle Carousel

This experiment combines the wonders of physics with the art of crafting, making it an enriching experience that ignites curiosity and fosters a deeper appreciation for the elegant dance of energy in our world.

7. Find Out if Water Conducts Electricity

In this captivating activity, students will explore the conductive properties of water and unlock the secrets of how electrical currents flow through different substances.

Learn more: Rookie Parenting

8. Roller Coaster Stem Experiment

By experimenting with various designs and track configurations, students will refine their problem-solving skills and gain valuable insights into the practical applications of physics and engineering.

Learn more: STEM Project

9.柠檬Battery

Engaging in this experiment not only teaches the basics of electrical circuits but also sparks curiosity about the natural world and the science behind it.

Learn more: Coffee Cups and Crayons

10. Watering Plants Using Different Liquids

我发现植物水合的奇迹ntriguing high school science experiment – “Watering Plants Using Different Liquids.” In this captivating project, students explore how various liquids impact plant growth and health.

Learn more: Lemon Lime Adventures

11. Measure Electrolytes Found in Sports Drinks

By conducting a series of tests and analyses, students will quantify the electrolyte content present in various sports drinks.

12. Relight the Flame Without Directly Touching It

这个迷人的项目challenges students to delve into the intriguing properties of heat transfer and combustion.

By exploring different methods to reignite a candle flame without physical contact, students will uncover the secrets of heat conduction, convection, and radiation.

Learn more: Stevespangler

13. Conduct Fingerprint Analysis

This captivating project immerses students in the intriguing world of crime scene investigations, where they will uncover the uniqueness of fingerprints and their role in forensic science.

14. Separate Water Into Hydrogen And Oxygen Using Electrolysis

This electrifying project allows students to explore the fascinating world of electrolysis and the decomposition of water into its elemental components.

Learn more: Navigating by Joy

15. Simple Color Detection Circuit

This experiment not only introduces fundamental concepts in electronics and circuitry but also opens up endless possibilities for real-life applications, from automated sorting systems to color-sensitive devices.

16. Carbon Sugar Snake

This enchanting project allows students to witness a dazzling display of science as they combine common household ingredients to create a dark, coiling “snake” made of carbon.

Learn more: Kiwi Co

17. Build a Hydraulic Elevator

This captivating project invites students to delve into the world of engineering and fluid mechanics. By constructing a working model of a hydraulic elevator, students will explore the principles of Pascal’s law and the fascinating concept of fluid pressure.

Learn more: Teach Beside Me

18. Brew up Some Root Beer

This enticing project invites students to explore the fascinating world of chemistry and fermentation while creating their own delicious and bubbly concoction.

Learn more: Home School Creations

19. Extracting Bismuth From Pepto-Bismol Tablets

This hands-on experiment not only sheds light on the principles of chemistry and lab techniques but also highlights the real-world applications of bismuth in medicine and various industries.

Learn more: Popscie

20. Solar-Powered Water Desalination

By designing and building a solar-powered water desalination system, students will learn how to harness the sun’s energy to purify saltwater and make it safe for consumption.

21. Applying Hooke’s Law: Make Your Own Spring Scale

By designing and constructing their very own spring scale, students will uncover the principles of Hooke’s Law and the relationship between force and displacement in a spring system.

22. Homemade Hand Warmer

By creating their own hand warmers using safe and easily accessible materials, students will witness the magic of heat generation through chemical processes.

Learn more: Steve Spangler

23. Explore the Concept of Symbiosis Involving Nitrogen-Fixing Bacteria.

By investigating how certain plants form a mutually beneficial bond with these bacteria, students will gain insights into the essential role of nitrogen fixation in the ecosystem.

Learn more: Education.com

24. Center of Gravity Experiment

This fascinating project invites students to explore the concept of the center of gravity and its role in determining stability.

25. Power up Homemade Batteries

This captivating project invites students to delve into the fascinating world of electrochemistry and energy generation.

Learn more: 123 Homeschool

26. Film Canister Explosions

Prepare for a blast of excitement and chemistry with the high school science experiment – “Film Canister Explosions!” This thrilling project invites students to explore the fascinating world of chemical reactions and pressure build-up.

27. Investigating Osmosis with Potato Slices

This hands-on experiment not only provides a practical understanding of osmosis but also highlights its relevance in everyday life, from understanding plant hydration to food preservation techniques.

28. Make Homemade Fly Trap

Delve into the fascinating world of insects with the high school science experiment – “Make Homemade Fly Trap!” This captivating project invites students to explore the principles of pest control and observe the behavior of flies.

29. Hydroponics: Gardening Without Soil

This exciting project invites students to explore innovative agricultural practices that harness water and nutrient solutions to grow plants.

By setting up their hydroponic system and nurturing plants through this method, students will witness the fascinating dynamics of root development and nutrient absorption.

30. Clothespin Airplane

As they test and modify their creations, students will delve into the principles of lift, thrust, and drag, gaining a deeper understanding of how these forces come together to keep airplanes soaring through the skies.

Learn more: Steamsational

Similar Posts:

• 37 Water Science Experiments: Fun & Easy
• Top 50 Fun Food Science Experiments
• 前40名有趣的乐高科学实验
• 40 Best Science Experiments & Projects for Middle School
• Top 30 Biology Experiments for High-School

Leave a Comment Cancel reply

Save my name and email in this browser for the next time I comment.

18 Must-Try Science Experiments for High School: From Basic Chemistry to Complex Reactions

Learners of all ages are enamored with scientific experiments:

P5 have been looking at changes of state in science, and today investigated the water cycle! We did an experiment with water & food colouring in a plastic bag to see if we could see any changes, and noticed signs of evaporation and condensation inside the bag @SLC_RAiSE #Science pic.twitter.com/cla3opitiT — Burgh Primary School (@BurghPrimary) October 25, 2023

This article will equip high school teachers with an arsenal of exciting science experiments that will keep their students engaged and learning. Offering projects across a variety of disciplines, from physics to biology, this carefully curated list will be suitable for learners at any level. By incorporating these experiments into their lesson plans, educators will be providing their students with valuable hands-on experience that complements their textbook knowledge. With easy-to-follow instructions and materials that are easily accessible, teaching science has never been more enjoyable!

Experiment 1: Investigating Osmosis with Potato Slices

This accompanying video offers a visual guide on how this osmosis project is conducted using potatoes. By the end, students will have a vivid understanding of osmotic movement and its effects.

Experiment 2: Making a Homemade Volcano

High school students have a wonderful opportunity to step into the shoes of a scientist with this exciting and educational experiment. They can construct their very own volcanic eruption, right from the safety of their classroom or home! By synergizing baking soda with vinegar, students will get a firsthand view of a thrilling chemical reaction that mimics the grandeur of a volcanic eruption. Beyond the sheer fun and spectacle, this experiment serves as an enlightening experience, imparting deeper insights into the complex world of chemical reactions.

Experiment 3: Exploring Density with Oil and Water

Experiment 4: building a simple electric motor.

High school students possess an innate curiosity, constantly seeking to understand the world around them. Dive deep into the captivating realm of electromagnetism with this enlightening project, revealing the intricate process that enables an electric motor to effortlessly transform electrical impulses into tangible mechanical movements. As students embark on this hands-on journey, they’ll gain an intimate appreciation for the underlying principles that power much of today’s technology.

Experience the mesmerizing magnificence of an electric motor as this video unravels the mystery behind its seamless conversion of electrical energy into mechanical power. Unlock the inner workings of this wonder machine in the science projects for high school.

Experiment 5: Testing Acids and Bases with Red Cabbage

Experiment 6: observing microorganisms with a microscope, experiment 7: studying chemical reactions with alka-seltzer experiment, experiment 8: measuring the speed of light with a microwave oven, experiment 9: demonstrating newton’s third law of motion with balloons, experiment 10: observing the greenhouse effect with sunlight and jars, experiment 11: investigating chromatography with markers, experiment 12: creating a simple electromagnet, experiment 13: examining photosynthesis with leaf disks, experiment 14: extracting dna from strawberries, experiment 15: building a mini tesla coil, additional 3 fun science experiments for high school, experiment 16: making invisible ink with lemon juice, experiment 17: creating rainbow fire with salt, experiment 18: exploring bioluminescence with glowing bacteria, useful science experiments resources, leave a comment cancel reply.

• Skip to primary navigation
• Skip to main content
• Skip to primary sidebar

• FREE Experiments
• Kitchen Science
• Climate Change
• Egg Experiments
• Fairy Tale Science
• Edible Science
• Human Health
• Inspirational Women
• Forces and Motion
• Science Fair Projects
• STEM Challenges
• Science Sparks Books
• Contact Science Sparks
• Science Resources for Home and School

Top 5 physics experiments you can do at home

October 17, 2022 By Emma Vanstone Leave a Comment

Physics is key to understanding the world around us. While some aspects may seem tricky to understand, many fundamental physics concepts can be broken down into simple concepts, some of which can be demonstrated using basic equipment at home.

This list of 5 physics experiments you can try at home is a great starting point for understanding physics and, hopefully a source of inspiration for little scientists everywhere!

Physics experiments you can do at home

1. archimedes and density.

The story behind Archimedes’ discovery of density is that he was asked by the King of Sicily to work out whether a goldsmith had replaced some gold from a crown with silver. Archimedes needed to determine if the goldsmith had cheated without damaging the crown.

The crown weighed the same as the gold the King had given the goldsmith, but gold is more dense than silver, so if there were silver in the crown its density would be less than if it were pure gold. Archimedes realised that if he could measure the crown’s volume, he could work out its density, but calculating the volume of a crown shape was a tough challenge. According to the story, Archimedes was having a bath one day when he realised the water level rose as he lowered himself into the bathtub. He realised that the volume of water displaced was equal to the volume of his body in the water.

Archimedes placed the crown in water to work out its density and realised the goldsmith had cheated the king!

Density Experiment

One fun way to demonstrate density is to make a density column. Choose a selection of liquids and place them in density order, from the most dense to the least dense. Carefully pour a small amount of each into a tall jar or glass, starting with the most dense. You should end up with a colourful stack of liquids!

2. Split light into the colours of the rainbow

Isaac Newton experimented with prisms and realised that light is made up of different colours ( the colours of the rainbow ). Newton made this discovery in the 1660s. It wasn’t until the 1900s that physicists discovered the electromagnetic spectrum , which includes light waves we can’t see, such as microwaves, x-ray waves, infrared and gamma rays.

How to split light

Splitting white light into the colours of the rainbow sounds tricky, but all you need is a prism. A prism is a transparent block shaped so light bends ( refracts ) as it passes through. Some colours bend more than others, so the whole spectrum of colours can be seen.

If you don’t have a prism, you can also use a garden hose! Stand with your back to the sun, and you’ll see a rainbow in the water! This is because drops of water act like a prism.

3. Speed of Falling Objects

Galileo’s falling objects.

Aristotle thought that heavy objects fell faster than lighter objects, a theory later disproved by Galileo .

It is said that Galileo dropped two cannonballs with different weights from the leaning tower of Pisa, which hit the ground at the same time. All objects accelerate at the same rate as they fall.

If you drop a feather and a hammer from the same height, the hammer will hit the ground first, but this is because of air resistance!

If a hammer and feather are dropped somewhere with no air resistance, they hit the ground simultaneously. Commander David Scott proved this was true on the Apollo 15 moonwalk!

Hammer and Feather Experiment on the Moon

Brian Cox also proved Galileo’s theory to be correct by doing the same experiment in a vacuum!

While you won’t be able to replicate a hammer or heavy ball and feather falling, you can investigate with two objects of the same size but different weights. This means the air resistance is the same for both objects, so the only difference is the weight.

Take two empty water bottles of the same size. Fill one to the top with water and leave the other empty. Drop them from the same height. Both will hit the ground at the same time!

4. Newton’s Laws of Motion

Sir Isaac Newton pops up a lot in any physics book as he came up with many of the laws that describe our universe and is undoubtedly one of the most famous scientists of all time. Newton’s Laws of Motion describe how things move and the relationship between a moving object and the forces acting on it.

Making and launching a mini rocket is a great way to learn about Newton’s Laws of Motion .

The rocket remains motionless unless a force acts on it ( Newton’s First Law ).

The acceleration of the rocket is affected by its mass. If you increase the mass of the rocket, its acceleration will be less than if it had less mass ( Newton’s Second Law ).

The equal and opposite reaction from the gas forcing the cork downwards propels the rocket upwards ( Newton’s Third Law ).

4. Pressure

Pressure is the force per unit area.

Imagine standing on a Lego brick. If you stand on a large brick, it will probably hurt. If you stand on a smaller brick with the same force it will hurt more as the pressure is greater!

Snowshoes are usually very wide. This is to reduce the pressure on the snow so it sinks less as people walk on it.

Pressure and Eggs

If you stand on one egg, it will most likely break. If you stand on lots of eggs with the same force, you increase the area the force is applied over and, therefore, reduce the pressure on each individual egg.

That’s five easy physics experiments you can do at home! Can you think of any more?

Last Updated on June 14, 2024 by Emma Vanstone

Safety Notice

Science Sparks ( Wild Sparks Enterprises Ltd ) are not liable for the actions of activity of any person who uses the information in this resource or in any of the suggested further resources. Science Sparks assume no liability with regard to injuries or damage to property that may occur as a result of using the information and carrying out the practical activities contained in this resource or in any of the suggested further resources.

These activities are designed to be carried out by children working with a parent, guardian or other appropriate adult. The adult involved is fully responsible for ensuring that the activities are carried out safely.

Reader Interactions

Leave a reply cancel reply.

Your email address will not be published. Required fields are marked *

• Grades 6-12
• School Leaders

Have you seen our latest free teacher workshop?

72 Easy Science Experiments Using Materials You Already Have On Hand

Because science doesn’t have to be complicated.

If there is one thing that is guaranteed to get your students excited, it’s a good science experiment! While some experiments require expensive lab equipment or dangerous chemicals, there are plenty of cool projects you can do with regular household items. We’ve rounded up a big collection of easy science experiments that anybody can try, and kids are going to love them!

Easy Chemistry Science Experiments

Easy physics science experiments, easy biology and environmental science experiments, easy engineering experiments and stem challenges.

1. Taste the Rainbow

Teach your students about diffusion while creating a beautiful and tasty rainbow! Tip: Have extra Skittles on hand so your class can eat a few!

Learn more: Skittles Diffusion

2. Crystallize sweet treats

Crystal science experiments teach kids about supersaturated solutions. This one is easy to do at home, and the results are absolutely delicious!

Learn more: Candy Crystals

3. Make a volcano erupt

This classic experiment demonstrates a chemical reaction between baking soda (sodium bicarbonate) and vinegar (acetic acid), which produces carbon dioxide gas, water, and sodium acetate.

Learn more: Best Volcano Experiments

4. Make elephant toothpaste

This fun project uses yeast and a hydrogen peroxide solution to create overflowing “elephant toothpaste.” Tip: Add an extra fun layer by having kids create toothpaste wrappers for plastic bottles.

5. Blow the biggest bubbles you can

Add a few simple ingredients to dish soap solution to create the largest bubbles you’ve ever seen! Kids learn about surface tension as they engineer these bubble-blowing wands.

Learn more: Giant Soap Bubbles

6. Demonstrate the “magic” leakproof bag

All you need is a zip-top plastic bag, sharp pencils, and water to blow your kids’ minds. Once they’re suitably impressed, teach them how the “trick” works by explaining the chemistry of polymers.

Learn more: Leakproof Bag

7. Use apple slices to learn about oxidation

Have students make predictions about what will happen to apple slices when immersed in different liquids, then put those predictions to the test. Have them record their observations.

Learn more: Apple Oxidation

8. Float a marker man

Their eyes will pop out of their heads when you “levitate” a stick figure right off the table! This experiment works due to the insolubility of dry-erase marker ink in water, combined with the lighter density of the ink.

Learn more: Floating Marker Man

9. Discover density with hot and cold water

There are a lot of easy science experiments you can do with density. This one is extremely simple, involving only hot and cold water and food coloring, but the visuals make it appealing and fun.

Learn more: Layered Water

10. Layer more liquids

This density demo is a little more complicated, but the effects are spectacular. Slowly layer liquids like honey, dish soap, water, and rubbing alcohol in a glass. Kids will be amazed when the liquids float one on top of the other like magic (except it is really science).

Learn more: Layered Liquids

11. Grow a carbon sugar snake

Easy science experiments can still have impressive results! This eye-popping chemical reaction demonstration only requires simple supplies like sugar, baking soda, and sand.

Learn more: Carbon Sugar Snake

12. Mix up some slime

Tell kids you’re going to make slime at home, and watch their eyes light up! There are a variety of ways to make slime, so try a few different recipes to find the one you like best.

13. Make homemade bouncy balls

These homemade bouncy balls are easy to make since all you need is glue, food coloring, borax powder, cornstarch, and warm water. You’ll want to store them inside a container like a plastic egg because they will flatten out over time.

Learn more: Make Your Own Bouncy Balls

14. Create eggshell chalk

Eggshells contain calcium, the same material that makes chalk. Grind them up and mix them with flour, water, and food coloring to make your very own sidewalk chalk.

Learn more: Eggshell Chalk

15. Make naked eggs

This is so cool! Use vinegar to dissolve the calcium carbonate in an eggshell to discover the membrane underneath that holds the egg together. Then, use the “naked” egg for another easy science experiment that demonstrates osmosis .

Learn more: Naked Egg Experiment

16. Turn milk into plastic

This sounds a lot more complicated than it is, but don’t be afraid to give it a try. Use simple kitchen supplies to create plastic polymers from plain old milk. Sculpt them into cool shapes when you’re done!

17. Test pH using cabbage

Teach kids about acids and bases without needing pH test strips! Simply boil some red cabbage and use the resulting water to test various substances—acids turn red and bases turn green.

Learn more: Cabbage pH

18. Clean some old coins

Use common household items to make old oxidized coins clean and shiny again in this simple chemistry experiment. Ask kids to predict (hypothesize) which will work best, then expand the learning by doing some research to explain the results.

Learn more: Cleaning Coins

19. Pull an egg into a bottle

This classic easy science experiment never fails to delight. Use the power of air pressure to suck a hard-boiled egg into a jar, no hands required.

Learn more: Egg in a Bottle

20. Blow up a balloon (without blowing)

Chances are good you probably did easy science experiments like this when you were in school. The baking soda and vinegar balloon experiment demonstrates the reactions between acids and bases when you fill a bottle with vinegar and a balloon with baking soda.

21 Assemble a DIY lava lamp

This 1970s trend is back—as an easy science experiment! This activity combines acid-base reactions with density for a totally groovy result.

22. Explore how sugary drinks affect teeth

The calcium content of eggshells makes them a great stand-in for teeth. Use eggs to explore how soda and juice can stain teeth and wear down the enamel. Expand your learning by trying different toothpaste-and-toothbrush combinations to see how effective they are.

Learn more: Sugar and Teeth Experiment

23. Mummify a hot dog

If your kids are fascinated by the Egyptians, they’ll love learning to mummify a hot dog! No need for canopic jars , just grab some baking soda and get started.

24. Extinguish flames with carbon dioxide

This is a fiery twist on acid-base experiments. Light a candle and talk about what fire needs in order to survive. Then, create an acid-base reaction and “pour” the carbon dioxide to extinguish the flame. The CO2 gas acts like a liquid, suffocating the fire.

25. Send secret messages with invisible ink

Turn your kids into secret agents! Write messages with a paintbrush dipped in lemon juice, then hold the paper over a heat source and watch the invisible become visible as oxidation goes to work.

Learn more: Invisible Ink

26. Create dancing popcorn

This is a fun version of the classic baking soda and vinegar experiment, perfect for the younger crowd. The bubbly mixture causes popcorn to dance around in the water.

27. Shoot a soda geyser sky-high

You’ve always wondered if this really works, so it’s time to find out for yourself! Kids will marvel at the chemical reaction that sends diet soda shooting high in the air when Mentos are added.

Learn more: Soda Explosion

28. Send a teabag flying

Hot air rises, and this experiment can prove it! You’ll want to supervise kids with fire, of course. For more safety, try this one outside.

Learn more: Flying Tea Bags

29. Create magic milk

This fun and easy science experiment demonstrates principles related to surface tension, molecular interactions, and fluid dynamics.

Learn more: Magic Milk Experiment

30. Watch the water rise

Learn about Charles’s Law with this simple experiment. As the candle burns, using up oxygen and heating the air in the glass, the water rises as if by magic.

Learn more: Rising Water

31. Learn about capillary action

Kids will be amazed as they watch the colored water move from glass to glass, and you’ll love the easy and inexpensive setup. Gather some water, paper towels, and food coloring to teach the scientific magic of capillary action.

Learn more: Capillary Action

32. Give a balloon a beard

Equally educational and fun, this experiment will teach kids about static electricity using everyday materials. Kids will undoubtedly get a kick out of creating beards on their balloon person!

Learn more: Static Electricity

33. Find your way with a DIY compass

Here’s an old classic that never fails to impress. Magnetize a needle, float it on the water’s surface, and it will always point north.

Learn more: DIY Compass

34. Crush a can using air pressure

Sure, it’s easy to crush a soda can with your bare hands, but what if you could do it without touching it at all? That’s the power of air pressure!

35. Tell time using the sun

While people use clocks or even phones to tell time today, there was a time when a sundial was the best means to do that. Kids will certainly get a kick out of creating their own sundials using everyday materials like cardboard and pencils.

Learn more: Make Your Own Sundial

36. Launch a balloon rocket

Grab balloons, string, straws, and tape, and launch rockets to learn about the laws of motion.

37. Make sparks with steel wool

All you need is steel wool and a 9-volt battery to perform this science demo that’s bound to make their eyes light up! Kids learn about chain reactions, chemical changes, and more.

Learn more: Steel Wool Electricity

38. Levitate a Ping-Pong ball

Kids will get a kick out of this experiment, which is really all about Bernoulli’s principle. You only need plastic bottles, bendy straws, and Ping-Pong balls to make the science magic happen.

39. Whip up a tornado in a bottle

There are plenty of versions of this classic experiment out there, but we love this one because it sparkles! Kids learn about a vortex and what it takes to create one.

Learn more: Tornado in a Bottle

40. Monitor air pressure with a DIY barometer

This simple but effective DIY science project teaches kids about air pressure and meteorology. They’ll have fun tracking and predicting the weather with their very own barometer.

Learn more: DIY Barometer

41. Peer through an ice magnifying glass

Students will certainly get a thrill out of seeing how an everyday object like a piece of ice can be used as a magnifying glass. Be sure to use purified or distilled water since tap water will have impurities in it that will cause distortion.

Learn more: Ice Magnifying Glass

42. String up some sticky ice

Can you lift an ice cube using just a piece of string? This quick experiment teaches you how. Use a little salt to melt the ice and then refreeze the ice with the string attached.

Learn more: Sticky Ice

43. “Flip” a drawing with water

Light refraction causes some really cool effects, and there are multiple easy science experiments you can do with it. This one uses refraction to “flip” a drawing; you can also try the famous “disappearing penny” trick .

Learn more: Light Refraction With Water

44. Color some flowers

We love how simple this project is to re-create since all you’ll need are some white carnations, food coloring, glasses, and water. The end result is just so beautiful!

45. Use glitter to fight germs

Everyone knows that glitter is just like germs—it gets everywhere and is so hard to get rid of! Use that to your advantage and show kids how soap fights glitter and germs.

Learn more: Glitter Germs

46. Re-create the water cycle in a bag

You can do so many easy science experiments with a simple zip-top bag. Fill one partway with water and set it on a sunny windowsill to see how the water evaporates up and eventually “rains” down.

Learn more: Water Cycle

47. Learn about plant transpiration

Your backyard is a terrific place for easy science experiments. Grab a plastic bag and rubber band to learn how plants get rid of excess water they don’t need, a process known as transpiration.

Learn more: Plant Transpiration

48. Clean up an oil spill

Before conducting this experiment, teach your students about engineers who solve environmental problems like oil spills. Then, have your students use provided materials to clean the oil spill from their oceans.

Learn more: Oil Spill

49. Construct a pair of model lungs

Kids get a better understanding of the respiratory system when they build model lungs using a plastic water bottle and some balloons. You can modify the experiment to demonstrate the effects of smoking too.

Learn more: Model Lungs

50. Experiment with limestone rocks

Kids  love to collect rocks, and there are plenty of easy science experiments you can do with them. In this one, pour vinegar over a rock to see if it bubbles. If it does, you’ve found limestone!

Learn more: Limestone Experiments

51. Turn a bottle into a rain gauge

All you need is a plastic bottle, a ruler, and a permanent marker to make your own rain gauge. Monitor your measurements and see how they stack up against meteorology reports in your area.

Learn more: DIY Rain Gauge

52. Build up towel mountains

This clever demonstration helps kids understand how some landforms are created. Use layers of towels to represent rock layers and boxes for continents. Then pu-u-u-sh and see what happens!

Learn more: Towel Mountains

53. Take a play dough core sample

Learn about the layers of the earth by building them out of Play-Doh, then take a core sample with a straw. ( Love Play-Doh? Get more learning ideas here. )

Learn more: Play Dough Core Sampling

54. Project the stars on your ceiling

Use the video lesson in the link below to learn why stars are only visible at night. Then create a DIY star projector to explore the concept hands-on.

Learn more: DIY Star Projector

55. Make it rain

Use shaving cream and food coloring to simulate clouds and rain. This is an easy science experiment little ones will beg to do over and over.

Learn more: Shaving Cream Rain

56. Blow up your fingerprint

This is such a cool (and easy!) way to look at fingerprint patterns. Inflate a balloon a bit, use some ink to put a fingerprint on it, then blow it up big to see your fingerprint in detail.

57. Snack on a DNA model

Twizzlers, gumdrops, and a few toothpicks are all you need to make this super-fun (and yummy!) DNA model.

Learn more: Edible DNA Model

58. Dissect a flower

Take a nature walk and find a flower or two. Then bring them home and take them apart to discover all the different parts of flowers.

59. Craft smartphone speakers

No Bluetooth speaker? No problem! Put together your own from paper cups and toilet paper tubes.

Learn more: Smartphone Speakers

60. Race a balloon-powered car

Kids will be amazed when they learn they can put together this awesome racer using cardboard and bottle-cap wheels. The balloon-powered “engine” is so much fun too.

Learn more: Balloon-Powered Car

61. Build a Ferris wheel

You’ve probably ridden on a Ferris wheel, but can you build one? Stock up on wood craft sticks and find out! Play around with different designs to see which one works best.

Learn more: Craft Stick Ferris Wheel

62. Design a phone stand

There are lots of ways to craft a DIY phone stand, which makes this a perfect creative-thinking STEM challenge.

63. Conduct an egg drop

Put all their engineering skills to the test with an egg drop! Challenge kids to build a container from stuff they find around the house that will protect an egg from a long fall (this is especially fun to do from upper-story windows).

Learn more: Egg Drop Challenge Ideas

64. Engineer a drinking-straw roller coaster

STEM challenges are always a hit with kids. We love this one, which only requires basic supplies like drinking straws.

Learn more: Straw Roller Coaster

65. Build a solar oven

Explore the power of the sun when you build your own solar ovens and use them to cook some yummy treats. This experiment takes a little more time and effort, but the results are always impressive. The link below has complete instructions.

Learn more: Solar Oven

66. Build a Da Vinci bridge

There are plenty of bridge-building experiments out there, but this one is unique. It’s inspired by Leonardo da Vinci’s 500-year-old self-supporting wooden bridge. Learn how to build it at the link, and expand your learning by exploring more about Da Vinci himself.

Learn more: Da Vinci Bridge

67. Step through an index card

This is one easy science experiment that never fails to astonish. With carefully placed scissor cuts on an index card, you can make a loop large enough to fit a (small) human body through! Kids will be wowed as they learn about surface area.

68. Stand on a pile of paper cups

Combine physics and engineering and challenge kids to create a paper cup structure that can support their weight. This is a cool project for aspiring architects.

Learn more: Paper Cup Stack

69. Test out parachutes

Gather a variety of materials (try tissues, handkerchiefs, plastic bags, etc.) and see which ones make the best parachutes. You can also find out how they’re affected by windy days or find out which ones work in the rain.

Learn more: Parachute Drop

70. Recycle newspapers into an engineering challenge

It’s amazing how a stack of newspapers can spark such creative engineering. Challenge kids to build a tower, support a book, or even build a chair using only newspaper and tape!

Learn more: Newspaper STEM Challenge

71. Use rubber bands to sound out acoustics

Explore the ways that sound waves are affected by what’s around them using a simple rubber band “guitar.” (Kids absolutely love playing with these!)

Learn more: Rubber Band Guitar

72. Assemble a better umbrella

Challenge students to engineer the best possible umbrella from various household supplies. Encourage them to plan, draw blueprints, and test their creations using the scientific method.

Learn more: Umbrella STEM Challenge

Plus, sign up for our newsletters to get all the latest learning ideas straight to your inbox.

You Might Also Like

Magic Milk Experiment: How-To Plus Free Worksheet

This classic experiment teaches kids about basic chemistry and physics. Continue Reading

Copyright © 2024. All rights reserved. 5335 Gate Parkway, Jacksonville, FL 32256

• Our Mission

How Any High School Science or Math Teacher Can Confidently Tackle Physics

While the idea of teaching physics without formal training in the subject can be daunting, it can be done with just a little help.

Physics teaching in the United States has a chicken-and-egg problem. Many districts and schools ( typically, diverse urban schools and rural schools ) do not offer the course or perhaps have a single section. Independent of the national teacher shortage, universities produce few physics teachers , with two-thirds of institutions producing none. In fact, according to the latest available data from the 2012–13 school year, fewer than half of physics teachers have any physics training, and only about a quarter have a degree in physics .

Without a good teacher, few students enroll in physics, and with low enrollment, a district has no need for a physics teacher. As a result, many teachers get pulled into physics to cover a section. Furthermore, adoption of the Next Generation Science Standards (NGSS) has required more teachers at every grade level to cover physics concepts in a more robust way than before. While the mere idea of physics may cause some people high anxiety, the reality is that with the right support, any high school science or math teacher can teach physics.

Real particle physics analysis by UK secondary school students using the ATLAS Open Data: an illustration through a collection of original student research

• Regular Article
• Open access
• Published: 02 September 2024
• Volume 139 , article number  781 , ( 2024 )

Cite this article

You have full access to this open access article

• Eimear Conroy   ORCID: orcid.org/0000-0002-0215-2767 1 ,
• Alan Barr 1   na1 ,
• Ynyr Harris 1   na1 ,
• Julie Kirk 2   na1 ,
• Emmanuel Olaiya 2   na1 &
• Richard Phillips 3   na1  

Since the 2020 release of \(10 \hbox { fb}^{-1}\) of integrated luminosity of proton–proton collision data to the public by the ATLAS experiment, significant potential for its use for youth engagement in physics and citizen science has been present. In particular, this article aims to address whether, if provided adequate training and resources, high school students are capable of leveraging the ATLAS Open Data to semi-autonomously develop their own original research projects. To this end, a repository of interactive Python Jupyter notebook training materials was developed, incrementally increasing in difficulty; in the initial instalments no prior knowledge of particle physics or Python coding is assumed, while in the latter stages students emulate the steps of a real Higgs boson search using ATLAS data. This programme was implemented in secondary schools throughout the UK during the 2022/23 academic year and is presented in this article through a collection of research projects developed by a selection of participating students.

Similar content being viewed by others

Sharing LHC Research and Discovery with High School Students

The Physics of Everyday Life Toolbox for Basic Physics Courses

Bringing computational science to the public

Explore related subjects.

• Artificial Intelligence

Avoid common mistakes on your manuscript.

During the 2022/23 academic year, a repository of interactive Python Jupyter notebook training materials for meaningfully interacting with the ATLAS Open Data were rolled out to schools across the UK by the UK-based Institute for Research in Schools (IRIS), an organisation which aims to promote original research performed by secondary school students. This is accomplished by connecting and supporting schools and teachers with projects compiled by academic and industry partners, spanning topics from ionic liquid chemistry to Earth observation. This ATLAS Open Data programme and set of resources were developed over the span of three years, including alpha and beta tests with groups of students participating in the International Particle Physics Masterclass at the University of Oxford and with a pilot of approximately 50 students in 6 UK schools in the 2021/22 academic year, respectively.

The full programme was rolled out to approximately 300 participating students from 25 participating schools nationwide in the 2022/23 academic year, where students worked through the repository of notebooks semi-autonomously in small groups, supported by their teachers, who were in turn supported by IRIS and a team of particle physics researchers at the University of Oxford and the Rutherford Appleton Laboratory (RAL). A map illustrating the distribution of participating schools across the United Kingdom is shown in Fig. 1 . In general, schools participated for several months to one full academic year. The time invested per week was at the discretion of individual schools and teachers; an example of a common arrangement was a weekly after-school club. With each notebook incrementally increasing in difficulty, the decision was left to the students and their teachers, after which notebook they wished to consolidate their learning and produce an original research project.

A map illustrating the distribution of all UK schools participating in the IRIS ATLAS Open Data project in the 2022/23 academic year

The intended research output of these original student projects was a set of posters, with a subset to be presented at one of the IRIS Student Conferences in summer 2023. The target audience of these conferences are the students participating in every IRIS-supported project in a given year, of which the ATLAS Open Data project is one. These conferences allow students to experience an academic conference setting, to present talks and posters to their peers, and ask to questions of their peers’ work in turn. Additionally, following the conference, participating student groups are invited to write proceedings.

This article presents the collection of conference proceedings prepared by participating students in the ATLAS Open Data project, structured such that the order in which the students’ projects are presented approximately the topic order of the repository of notebooks provided to them. All work and writing in Sects.  2 – 8 is the students’ own, with the exception of framing prefaces written in italics and light-touch editing for legibility.

The contributed pieces represent solely the work and perspectives of the secondary school students by whom the individual works were produced, and do not represent the viewpoints of the ATLAS Collaboration.

2 Introduction

To date, ATLAS experiment [ 1 ] at the Large Hadron Collider [ 2 ] has released a total integrated luminosity of 10 \(\hbox {fb}^{-1}\) of \(\sqrt{s}=13\) TeV proton–proton ( pp ) collision data and corresponding Monte Carlo (MC) simulations [ 3 ] to the public, in addition to \(1\hbox { fb}^{-1}\) of \(\sqrt{s}=8\) TeV pp collision data and corresponding MC [ 4 ], in accordance with the ATLAS Open Data Access policy [ 5 ]. Alongside the release of the data, provided in ROOT format [ 6 ], the ATLAS Collaboration has provided extensive documentation for the public datasets and a variety of tools for their usage, targeting secondary school students, undergraduates, graduate students, and the teachers and lecturers who supervise them. One such example is a collection of Jupyter notebooks [ 7 ] provided on the ATLAS Open Data portal [ 8 ], containing examples of Python data analysis activities and, exploiting the visual advantages of the Jupyter notebook format, accompanying text and images.

In this project, we build on the collection of Jupyter notebooks provided by ATLAS to create a repository [ 9 ] of training materials targeting secondary school students in the UK, adding notebooks introducing students to python coding, comprehensive background information for each exercise, extensively commenting example code, integrated exercises with solutions for students, and prompts for learning extension exercises which may also serve as ideas for independent student projects. We aim to address whether, if provided the correct training and resources, secondary school students can produce original particle physics research using the ATLAS Open Data. The created repository comprises a set of seven Jupyter notebooks and supplementary materials, with no prior knowledge of Python coding or particle physics assumed. We take a scaffolded approach; each notebook becomes incrementally more challenging than the previous. The topics covered by each notebook are shown in Fig. 2 .

The structure and topics covered by the repository of training materials created for UK secondary school students

The structure of each individual notebook is as follows:

Introduction of learning goals;

Necessary background information presented in a variety of formats—text, diagram, video;

Worked data analysis example, including accompanying text and substantively commented code;

“Try it yourself” exercise. Particularly in early notebooks, the code structure is provided, with blanks left to be filled by students;

At the conclusion of each notebook, prompts for learning consolidation exercise/ independent project ideas are provided. These “off ramps” are chosen such that they may be completed using only the skills acquired to this point, providing flexibility to students and teachers with respect to the time and resources they wish to invest. Sections  2 – 8 present reports of these independent projects in the students’ own words, with the order selected to roughly follow the right-hand (blue) column of Fig.  1 so that this article reflects the structure of the repository of training materials.

3 Developing python coding skills to model projectile motion and analyse data from high-energy collisions in the Large Hadron Collider

By students of Limavady Grammar School : Leo Collins, Darcy Cooper, Ella Feeney, Callum Gilpin, Emily Harnett, Annabelle Hunter, Norman Ling, Rebecca McCausland, Aoife McLaughlin, Olivia McLernon, Ryan Wilson.

3.1 Preface

In this section, the students consolidate skills developed in the first four notebooks of the repository. They rely on background information on the Large Hadron Collider and the ATLAS experiment developed from notebook 1, Python coding skills developed in notebook 2, and the ability to interact with ROOT files to produce histograms of event kinematics developed in notebook 3. They employ these skills to develop a Python model of project motion and compare the benefits of their model with analytical methods learned in school, extract and plot lepton multiplicity in a sample of simulated events, and to reproduce the Z boson mass peak.

3.2 Summary

In this project, we developed skills in coding with Python. Using these skills, we were able to write the code necessary to solve projectile motion questions. This provided us with an insight into the advantages and disadvantages of using this method over completing these calculations manually. We then used our Python skills to create histograms via coding. The histograms produced show that, in the chosen samples of simulated data with no selections applied, it is most likely that only one lepton will be reconstructed in each event inside the Large Hadron Collider. Finally, we applied appropriate selections to the data to reconstruct the Z boson invariant mass peak. Via the use of Python code, we determined the mass of the Z boson to be approximately equal to 90 GeV.

To use Python programming language to write a programme to find the distance travelled by a 10 kg projectile fired at 15 m/s at \(45^\circ\) above the horizontal from a point 2 m above the ground;

To use Python programming language to display data in the form of a histogram of simulated high-energy collisions in the Large Hadron Collider;

To use Python to analyse data from high-energy collisions in the Large Hadron Collider and determine the mass of the Z boson.

3.4 Background information

3.4.1 python coding.

Python is a popular general-purpose programming language that can be used for a wide variety of applications. It is used to build websites, software, and perform analysis [ 10 ]. Python also offers the ability to easily automate processes through scripting, making it key for software testing, troubleshooting, and bug tracking. It plays a key role in data science tasks and is used to perform complex statistical calculations, visualise data, and create machine learning algorithms [ 11 ].

Python is used today in research to solve many of the world’s complex modern physics problems. However, it may also be applied to more basic physics equations, for example, calculations involving constant velocity. This can be done by simulating the motion of an object. If we know the object’s x positional coordinate at a particular time t and its instantaneous velocity v along the x -axis at that time, this will allow us to find the object’s x position a small time later \(\Delta t\) by substituting these values equation:

Similarly, Python can be used to solve basic problems of constant acceleration, by updating Eq.  1 to reflect that, since the object is now accelerating, its velocity is changed at every step and must be updated accordingly:

3.4.2 Projectile motion

Projectile motion is the motion of an object thrown or projected into the air, experiencing only the acceleration due to the force of gravity. The object is called a projectile, and its path is called its trajectory [ 12 ]. The resultant path is in effect a combination of two motions—horizontal and vertical. This allows us to apply the equations of motion separately in each orthogonal direction. Particles in a projectile follow a curved path known as a parabola.

3.4.3 The Large Hadron Collider

The Large Hadron Collider (LHC) [ 2 ] is the world’s largest and most powerful particle accelerator. It consists of a 27-kilometre ring of superconducting magnets with a number of accelerating ‘radio frequency cavity’ structures to boost the energy of the particles along the way [ 13 ].

3.4.4 ATLAS

The ATLAS detector [ 1 ] is a general-purpose particle physics experiment at the LHC at CERN. ATLAS is the largest detector of its kind and is designed to record the high-energy particle collisions of the LHC, which take place at a rate of 40 million interactions per second in the centre of the detector. The ATLAS Collaboration is a large global collaboration with the common aim of better understanding the fundamental constituents of matter and their interactions [ 14 ]. The ATLAS project investigates a wide variety of fundamental particles, from the Higgs boson to what makes up dark matter [ 15 ].

3.4.5 Fundamental particles

There are two types of fundamental particles, quarks and leptons. Each of these groups have six particles, related in pairs. The six quarks are paired into three generations; the up and down quark, the charm and the strange quark; and the top and bottom quark. The six leptons are also arranged into three generations; electrons and electron neutrinos, muons and muon neutrinos, and the tau and tau neutrino [ 16 ]. The electron, muon and tau lepton all have an electric charge and a sizeable mass, whereas the neutrinos are electrically neutral and have very small mass. There are four fundamental forces, three of which result from the exchange of particles called bosons. The W and Z bosons are responsible for the weak force [ 16 ].

The Z boson is very unstable and does not live long enough to be detected, so to find the Z boson we reconstruct it from its decay products. We will reconstruct Z bosons which have decayed into two leptons. To conserve charge and lepton number these leptons need to have an opposite charge and the same flavour, meaning we will be looking for a muon and an antimuon or an electron and an antielectron (positron).

3.5.1 Method

To code the simulation for a projectile motion, we used the following method:

Set the initial conditions of the projectile;

Make a loop over time steps;

In the loop, update the velocity of the projectile (only its y-coordinate changes).

3.5.2 Results

Graph showing the projectile x position against time (left) and y position against time (right)

From Fig.  3 , it can be determined that the ball hits the ground, −2 m below its starting position, at time t= 2.80 s. At this time, it is at a horizontal distance of 22.1 m from its starting point, moving with a vertical velocity of − 14.7 m/s and a horizontal velocity of 7.9 m/s.

3.5.3 Analysis and conclusion

Advantages of simulating projectile motion using Python:

Good method of graphically displaying projectile motion;

Efficient technique for analysing large amounts of data;

Once the code is correct there is no room for human error;

Provides a clear representation of calculations.

Disadvantages of simulating projectile motion using Python:

Requires a comprehensive understanding of Python code;

May be difficult to apply to more complex physics problems.

3.6.1 Method

Load a ROOT file containing MC simulation from the ATLAS Open Data [ 17 ] database by using the Python uproot library;

To fill histogram, extract the number of leptons from our TTree using uproot . A TTree is a container that keeps track of the information from a collision event. Fill the histogram using the .fill() function from the Python hist module;

Plot the histogram using the .plot() and plt.show() functions from the Python matplotlib library;

Title the histogram and create a labelled x -axis.

Normalise the histogram so that it shows the proportion of each number of leptons produced, up to a maximum value of 1, instead of the absolute number of collision events that produced the different numbers of leptons.

3.6.2 Results

Figure 4 shows the lepton multiplicity of simulated events accessed as described above. The strong peak around at 1 shows that the majority of open data events included have one lepton, a result which was expected given that the histogram was produced from a sample with a one-lepton inclusive filter applied.

A histogram showing the absolute number of leptons produced per collision event (left) and the same histogram normalised to 1 (right)

3.6.3 Analysis and conclusion

An advantage of displaying large datasets in histogram format is that the information can be viewed in a clear, concise way, allowing for the major features of the distribution of the data to be seen. From the MC simulation analysed, we were able to see trends in many leptons were produced in each collision event simulated in that sample. The histograms produced show that, most commonly, only 1 lepton is produced per collision event.

3.7.1 Method

First, open a ROOT file of data collisions file using uproot , and inspect the contents of the file;

Use the .arrays method of uproot to import only specific variables for each event, and then define a histogram, with x -axis named mass/GeV;

Make cuts in the data [ 17 ], requiring two leptons of the same flavour, and then cut the data again requiring that those two leptons are oppositely charged, to reconstruct the Z boson invariant mass;

Import the Python matplotlib module and plot the histogram.

3.7.2 Results

Figure  5 shows the mass of the particles fitting the above criteria. The strong peak at around 90GeV shows that this is the mass of the Z boson.

Graph showing the mass distribution of a Z boson

3.7.3 Analysis and conclusion

The \(qq\rightarrow Z\rightarrow ll\) process is not the only way in which Z can be produced at the LHC; it is possible for virtual interactions between quarks and antiquarks to produce two Z bosons which then both decay in the same way as mentioned above to create a final state with four leptons. Exploring this interaction could be an alternative method to determine the mass of the Z boson.

4 An investigation in to energy conservation in the decay of the Z boson

By students of Lady Manners School : Caleb Byrne, George Colver.

4.1 Preface

In this section, students consolidate skills developed in notebooks 4 and 5, reproducing Z boson mass peaks in events with two or four leptons in the final state through the application of object and event selections and reconstructing the four-momentum of pairs of leptons. The students also explore the impact of statistical concepts, such as sample size and bin width, on particle physics results.

4.2 Summary

In this article, we discuss our processes for reconstructing the mass of a Z boson by analysing the four-momentum of the decay products. This has been achieved by analysing events from the ATLAS Open Data [ 17 , 18 ], recorded (and simulated) by the ATLAS experiment [ 1 ] CERN. A number of filtering conditions were used to identify cases where Z bosons had decayed into lepton–antilepton pairs, and these events were analysed to determine the invariant mass of the original unstable particles. These evaluations were done using the Python programming language, and the results were presented as histograms that display the frequency density for a range of masses; the modal peak can be taken as the true mass of the Z boson.

4.3 Introduction

The ATLAS Experiment [ 1 ] is one of four experiments located at the Large Hadron Collider [ 2 ] (shown in Fig.  6 ), where protons are accelerated to relativistic speeds and collided together, producing a great number of particles, including the Z boson.

The Large Hadron Collider, a depiction of the four detectors present at the LHC: ATLAS, LHCb, ALICE and CMS [ 19 ]

The ATLAS Experiment consists of four main subsystems, each concerned with identifying and measuring the properties of different particles as they pass through. Moving outward from the collision point, these are:

The silicon Tracker shows the paths taken by charged particles, allowing the momentum of these particles to be calculated by analysing the curvature of their trajectories;

The Electromagnetic Calorimeter measures the energy of particles which interact electromagnetically, such as photons and electrons, by recording the electric signals produced by their passage through layers of liquid argon and dense absorber material;

The Hadronic Calorimeter measures the energy of strongly interacting hadrons using layers of steel in which the particles are stopped;

Finally the Muon Spectrometer measures the energy and momentum of muons, which pass through the previous layers interacting very little.

The layers described above are shown in Fig.  7 .

A slice of the different sub-detectors within the ATLAS detector [ 20 ]

As highlighted in Fig.  7 , particles cannot be detected until they reach the relevant layers of the detector. This can be a problem if a particle is unstable and prone to decay, like the Z boson, as it is not possible to detect directly. Instead, the particles that it decays into must be considered, and used to reconstruct the original particle.

The Z boson can decay in a few different ways [ 21 ]. The most common ( \(\sim\) 70% of decays) channel is the Z decay into hadrons; however, as these particles can be produced by a number of processes within the accelerator, this is not the cleanest possible signature to accurately reconstruct the Z boson. The second most common decay route ( \(\sim\) 20% of decays) is the decay into neutrinos. However, since neutrinos are incredibly weakly interacting, they can only be indirectly inferred from Missing Energy in an event. Finally the last decay route ( \(\sim\) 10% of decays) produces a lepton–antilepton pair, show in Fig.  8 , and this channel suits our criteria of being easily measurable, and a unique enough event to identify accurately.

A Feynman diagram of Z boson decaying into a lepton–antilepton pair

By also considering another channel, we can be yet more stringent in our selection of events. It is possible for two quarks to interact, exchanging a virtual particle, and both decaying into Z bosons which then decay via the channels already identified, as shown in Fig.  9 . This means that two different lepton–antilepton pairs are produced, each within the window of the Z boson mass, a more unlikely coincidence if they were produced by another process, allowing for a more confident identification of these events as being the decays of Z bosons.

Quark interaction, producing two unstable Z bosons

Once the criteria for an event containing a Z boson decay had been determined, we were able to use these events to reconstruct the mass of the Z boson.

Due to the laws of conservation of mass-energy and momentum, the total momentum and mass-energy of the lepton–antilepton pair produced must be equal to the total momentum and mass-energy of the Z Boson that produced them. We used a Lorentz vector to store these mass-energy/momentum values concisely, sum them, and finally calculated the invariant mass of the original Z Boson. The structure of a Lorentz vector is shown below:

The E component is the mass-energy of a particle, and the p components are the momentum in each of the three spatial directions. These vectors can be combined by simply adding the relevant elements, and once the resultant vector is known, the invariant mass can be calculated as shown below:

Using these tools, the invariant mass of a Z boson can now be calculated. We used a mixture of real, recorded events from the ATLAS Open Data [ 17 , 18 ], and simulated [ 22 ] events in our investigation to evaluate the accuracy of the simulations and ensure that our method works in both cases.

This was achieved by first loading the data (using the Python uproot library to load the files into ROOT TTree data structures from which the events could then be extracted), and then iterating through each event, selecting events where the number of leptons present was an even number greater than 0, where the number of positive (antileptons) and negative (leptons) leptons was identical and of the same flavour, and hence where a pair of leptons originating from a Z boson was produced. These pairs then had their Lorentz vectors filled and summed to allow the invariant mass of the original Z boson to be calculated. These values were then used to fill a histogram, ready for analysis and inspection.

4.5 Results

In order to improve the presentation of our results, we used 60 bins in our histograms, from 60 to 120 GeV. These results are discussed below.

Determining the mass of the Z boson in the two-lepton channel using both simulated data (left) and real data (right)

Initially, we determined the mass of the Z boson in the two-lepton channel, as shown in Fig.  10 . The modal mass peaks on both the left-hand plot, corresponding to simulated Z boson decays, and on the right-hand plot, corresponding to real data, suggest a mass for the Z boson of 91 GeV \(\pm 1\) GeV. Compared to the accepted result of 91.2 GeV, our measurement is in good agreement. The right-hand histogram containing real data events is very similar to the simulated data, suggesting a high degree of accuracy in the CERN simulations. The real data showed considerably more \(Z\rightarrow \mu \mu\) decays; however this is likely due to aspects of the particular data files we chose to analyse such as trigger thresholds or object filters, and not an actual discrepancy.

Determining the mass of the Z boson in the four-lepton channel using both simulated data (left) and real data (right)

Subsequently, we repeated our measurement of the Z boson mass in the four-lepton channel, as shown in Fig. 11 . For both the simulated (left) and real data (right) invariant mass distributions for the decays of two Z bosons to a total of four leptons are rather similar to those in the two lepton channel shown in Fig. 10 ; however, it is considerably less ‘clean’, tending to show more spikes at lower masses and unusual dips at \(\sim\) 88 GeV.

4.6 Analysis and conclusions

All histograms we generated from a large sample of data sets indicate a Z boson mass of \(\sim\) 90 GeV, which is very close to the accepted value of 91.2 GeV [ 21 ], giving credence to our methodology.

Upon reflection, we realised that the reason the distributions created by the events involving four leptons were less well-fitted to a bell curve was because the data sets containing these events that we were initially using were orders of magnitude smaller than the equivalent two lepton datasets; random errors and fluctuations were magnified, and the distribution formed by these events was subsequently far less smooth. Using a larger data set for decays producing four leptons produces a far smoother histogram, as shown in Fig.  12 .

Use of a large data set for decays producing 4 leptons produced a smoother histogram

4.7 Further investigation

Our research so far has provided us with some interesting and exciting results, and we plan to continue to investigate the concept of detecting particles by working backwards from their more stable decay products to determine information about the original, less stable particle. We plan to apply similar methods to those discussed here to study other particles such as high mass quarks, and other bosons such as the W and Higgs bosons. Looking at the theory behind more complex interactions and planning experimental procedures to glean information about them is also an exciting concept which we plan to explore in the future.

4.8 Acknowledgements

The research and findings presented in this article could not have occurred without the help of IRIS, the University of Oxford, RAL, Dr Neil Garrido (Regional Schools Engagement Lead), and Dr Ebbens (Lady Manners Head of Physics).

5 Investigation into experimental methods

By students of Lady Manners School : Eleanor Joinson, Robbie Milton.

5.1 Preface

In this section, students consolidate skills developed in notebooks 6 and 7, to search for the Higgs boson using two different methods; a bump hunt in the \(H\rightarrow \gamma \gamma\) channel, and a non-resonant search in the \(H\rightarrow WW\) channel. In the former, the students explore the concept that larger datasets lead to clearer signal peaks, while in the latter, ideas such as irreducible backgrounds and the use of simulation are explored.

5.2 Introduction

Peter Higgs and Francois Englert won the Nobel Prize for Physics in 2013 for their work on the Higgs boson. In 1964, Higgs submitted a paper which predicted the existence of the Higgs field, which allowed boson mass to be introduced to the Standard Model. To test for the Higgs field, the Large Hadron Collider (LHC) [ 2 ] was used to search for a Higgs ’particle’ associated with the Higgs field, which would be unstable. In 2012, the evidence gathered by the LHC at the ATLAS detector was sufficient and strong enough to officially ’discover’ the Higgs boson [ 23 , 24 ].

We decided to investigate how different statistical methods for searching for new particles are more convincing than others, and how the accuracy and validity of identical data can alter based upon the tests applied. The two tests we conducted were via the \(H\rightarrow \gamma \gamma\) channel, where the Higgs boson will decay into two photons resulting in a larger concentration of photons of the known mass of the Higgs boson, and the \(H\rightarrow WW\) channel which is tested using a non-resonant search, where the small amount of difference in the transverse masses [ 25 ] between the data and predicted backgrounds provide evidence for the Higgs boson. This allowed us to compare the two methods in terms of statistical significance so that the strength of the different tests could be evaluated.

5.3 Method 1

The method used to first prove the existence of the Higgs boson involved the \(H\rightarrow \gamma \gamma\) channel shown in Fig.  13 ; however, diphoton pairs very commonly produced in LHC collisions, which means that only after analysing billions of collisions can a clear ’bump’ in the otherwise continuous curve of the diphoton invariant mass produced. It is also impossible to know the exact collisions the Higgs boson was produced in, but the significant increase in the number of diphoton events around the mass of the Higgs boson is enough evidence to confidently prove its involvement.

\(H\rightarrow \gamma \gamma\) channel

Using the four diphoton data sets available in the ATLAS Open Data [ 26 ], each containing millions of events, we selected the relevant photon objects that met ’Tight’ requirements. Our requirements for a Tight photon were: Event passes photon trigger, photon object is reconstructed, photon has \(p_{\mathrm{T}}>\) 25 GeV, photon is in the ’central’ region of ATLAS ( \(|\eta |<\) 2.37), photon does not fall in the ’transition region’ of ATLAS (1.37 \(\le |\eta |\le\) 1.52) between the calorimeter barrel an endcap.

Once the good-quality photons were extracted, Lorentz vectors of their four-momenta were built, and their respective invariant masses were calculated. From each data set, we produced a histogram showing the diphoton invariant mass, shown in Fig. 14 .

Histograms showing diphoton invariant mass distributions for each of the four available data sets

These histograms were then combined to produce a single histogram showing the data from all sets in one graph, shown in the left-hand side of Fig.  15 . While not directly comparable due to differences in event weighting, an example plot from a full \(H\rightarrow \gamma \gamma\) analysis is shown for illustrative purposes on the right-hand side of Fig. 15 .

Histogram showing diphoton invariant mass for the combined four data sets (left), and a full ATLAS measurement of the \(H\rightarrow \gamma \gamma\) channel [ 27 ] included for illustrative purposes (right)

The full data show a clear ‘bump’ in the data at 125 GeV when compared to the fourth-order polynomial, which shows the predicted trend without the Higgs boson. We see a similar ’bump’ in our histogram, but it is not as clear as the ATLAS results. However, we do still produce data with a deviation from the predicted data trend at the known mass of the Higgs boson. This suggests that while our data are less conclusive than the full ATLAS data, it still shows some evidence of the Higgs boson.

5.4 Method 2

In addition to the \(H\rightarrow \gamma \gamma\) channel, the \(H\rightarrow WW\) channel, shown in Fig.  16 is an alternative method used to prove the existence of the Higgs boson. This method is tested using a non-resonant search, where we investigate the difference between the background prediction (created using simulations) and the data [ 28 ], with any remaining events after the background is subtracted indicating the presence of the Higgs boson. This search therefore relies heavily on accurate simulations.

Feynman diagram of the Higgs boson decaying to 2 W bosons signal (left), and the Standard Model diboson background production from two quarks (right)

The simulated background must be scaled to ensure the sample size is in proportion to the number of data events recorded. Then, ’good leptons’ must be selected to improve the signal-to-background ratio of the events we selected. These leptons must pass ’Tight’ requirements: The lepton must be isolated and central ( \(|\eta |<\) 2.37). If the event shows exactly two leptons (indicative of a \(H\rightarrow WW\) decay, with each lepton originating from a leptonic W boson decay) then their Lorentz vectors are created and so the transverse masses [ 25 ] can be calculated (using the .Mt() function which came inbuilt with the training Jupyter notebooks) and plotted on a histogram, showing the frequency of different transverse mass values, shown in Fig.  17 (left).

We repeated this process for the predicted background data. From there, we subtracted the simulated background from the data, shown in Fig.  17 (right). This showed a clear excess of events, that can then be explained by the existence of the Higgs boson.

Histogram of the raw transverse mass frequencies of one data set without the removal of the background diboson production (left), compared to the remaining events after the simulated background data is removed (right). The resulting ‘left-over’ events are evidence for the presence of the Higgs boson

5.5 Analysis and conclusions

Both the experiments that we conducted resulted in viable evidence for the existence of the Higgs boson. In the first experimental method, the histogram produced does show a slight deviation from the 4th-order polynomial curve seen over the rest of the data in a similar place (at 125 GeV) to the full ATLAS data. The reduced clarity of the ’bump’ can be explained by the much smaller data set available for us to use, as the production of the Higgs boson is a rare event.

The second method we conducted uses simulated data to help identify the Higgs boson, by removing the ’expected’ events present, assuming no Higgs boson, from the ’actual’ events recorded to observe some events that were previously unaccounted for. Due to the use of simulations and scaled predicted results, the data might not give as conclusive evidence for the presence of the Higgs boson, as the results were not generated directly from raw data. This is in contrast to the first experimental method, in which all data points were filtered and selected from the raw data collected by ATLAS.

As a result, we concluded that the first method, the \(H\rightarrow \gamma \gamma\) channel, yielded more convincing results showing good agreement with the known mass of the Higgs boson, even with the comparatively smaller data set.

Moving on with our research, we plan to investigate other channels of Higgs boson decay, continuing to evaluate the evidence produced. From there, we would look to refine the tests further, with the aim of improving the validity of the results, and how convincing the evidence for the existence of the Higgs boson would be.

6 Searching for the Higgs Boson through the \(\gamma \gamma\) and \(W^+W^-\) decay channels

By students of City of London Academy Highgate Hill : Dan Trinder, Elani Ponnampalam, and Radhe Das.

6.1 Preface

In this section, students exploit techniques developed in notebook 4 ( \(Z\rightarrow ll\) search ), notebook 5 ( \(ZZ\rightarrow 4\,l\) search ), notebook 6 ( \(H\rightarrow \gamma \gamma\) search ) and notebook 7 ( \(H\rightarrow WW\) search ) to explore their original ideas in analysis code development, validation and optimisation. The Z boson mass peak is used to validate analysis code before searches for the Higgs boson are conducted. Additionally, the students develop original data caching and multithreading mechanisms to increase the efficiency of their data processing.

6.2 Summary

The primary focus of this study was to detect the Higgs boson through exploration of the \(H\rightarrow \gamma \gamma\) and \(H\rightarrow WW\) decay channels. Utilising the publicly released 13 TeV proton–proton collision data recorded by the ATLAS experiment [ 1 ], we employed Python modules and libraries such as NumPy , pickle and uproot to analyse over 10 million events. In addition to this, we developed two new modules: the rootFile module, which was created to combine data from various sources, and a multithreading module, which accelerated data processing. We refined our analysis code initially by testing it on a simulated sample Standard Model \(Z\rightarrow l^+l^-\) and \(ZZ\rightarrow l^+l^-l^+l^-\) decays in the ATLAS detector, provided by the ATLAS Open Data [ 18 , 28 ]. Our analysis of the Standard Model simulation relied on two foundational principles in physics: the conservation of invariant mass and the conservation of momentum. Following this, the \(H\rightarrow \gamma \gamma\) decay channel [ 26 ], was explored using a technique known as ‘bump hunting’, and for the and \(H\rightarrow WW\) decay channel [ 28 ], we conducted a non-resonant search. Despite rigorous analysis of the ATLAS Open Data to which we had access, our results were inconclusive; we did not detect the Higgs boson.

6.3 Introduction

In 2012, the ATLAS [ 1 ] and CMS [ 29 ] collaborations detected the Higgs boson at the Large Hadron Collider (LHC) [ 2 ], decades after it was first theorised in 1964 [ 30 ]. This discovery not only confirmed the existence of the Higgs field—which is responsible for a particle’s mass—but also marked the beginning of a continuing effort to understand its properties [ 31 ]. These endeavours may shed more light on phenomena like dark matter, which makes up most of the universe’s mass content but remains undetected [ 32 ]. This study aims to contribute to this understanding through exploration of two of the most common decay channels of the Higgs boson: \(H\rightarrow \gamma \gamma\) and \(H\rightarrow W^+W^-\) . We employ an iterative approach; refining our code as the study progresses, as well as sophisticated computational techniques such as multithreading to accelerate data analysis. With this research we aim to play a role in furthering scientific understanding of the Higgs boson.

Given the unstable nature of the Higgs boson, we focused on studying its decay products rather than the particle itself, due to its short lifetime. By utilising two fundamental laws in physics: the conservation of invariant mass [ 33 ] and the conservation of four-momentum, it was possible to reconstruct the Higgs boson from selected decay channels, and study it further.

Using the publicly released collection of 13 TeV proton–proton collision data recorded by the ATLAS Experiment, we began stage one of our validation process. The first step was to filter out events that contained more than two leptons, then to further reduce the data by eliminating pairs that did not have the same flavour and opposite charge. We then combined their individual four-momenta and used this to calculate their reconstructed invariant mass. At this stage of our research, we began the development of a more sophisticated file loading system—a prototype of our later rootFile module—and incorporated a data caching mechanism to reduce the impact of any computational errors that may have occurred and caused us to lose significant amounts of progress.

Following this, we modified our lepton event data filtration criteria to discard events that did not have four leptons. Similarly to the previous stage, we removed lepton pairs that did not have two leptons with opposite charge and same flavour. In addition to this, we implemented a for loop that matched and analysed every possible lepton pair, then selected the two pairs that, based on their invariant mass, were most likely to have originated from a Z boson decay and plotted them on the histogram.

After these two validation steps, we explored the \(H\rightarrow \gamma \gamma\) decay channel, employing Python code and ROOT Lorentz vectors to reconstruct the invariant masses of the decay products of the Higgs boson, and examining the resultant histogram to determine whether we had succeeded in detecting the Higgs boson. A ’bump’ around 125 GeV—the invariant mass of the Higgs boson—would have confirmed its presence. Considering that we were detecting photons in this stage, not leptons, we updated the filtering criteria. To be counted as a valid event, the photon had to pass ‘tight’ requirements, be well isolated, pass the photon trigger, and have sufficient transverse momentum. The invariant masses of photon pairs that that satisfied each of those requirements were plotted on a histogram along with error bars displaying the statistical uncertainty in our measurement. In addition to this, we did a data-driven estimate of the background diphoton events and fitted the resulting cubic function to our histogram to make the \(H\rightarrow \gamma \gamma\) events easier to see. We also created a multithreading module to speed up the data processing, after realising that the analysis of 10 million events would take a substantial amount of time and utilised the revised rootFile module to make the data extraction more efficient.

During our exploration of the \(H\rightarrow WW\) decay channel, we conducted a non-resonant search. This is done by plotting the transverse mass [ 25 ] of the \(W^+W^-\) for experimental data recorded by the ATLAS experiment, then subtracting the transverse mass of Monte Carlo simulations of background events coming from the SM WW diboson background production from two quarks and plotting the result on a histogram. W bosons cannot be detected directly in the ATLAS detector because they are unstable and decay too quickly. Instead, we look at their decay products—a lepton and missing energy from a neutrino—to confirm the W bosons’ presence. In addition to this, each simulated event had to be scaled to account for the greater number of events in the Monte Carlo sample than in the ATLAS data. After subtracting the simulated backgrounds, we plotted the resultant histogram, which displayed our measured Higgs signal.

6.5 Results

The focus of our study was the exploration of two common decay channels of the Higgs boson: \(H\rightarrow \gamma \gamma\) and \(H\rightarrow W^+W^-\) . Prior to investigating these decays, we carried out two stages of preliminary tests to ensure our code was reliable.

6.5.1 Code validation using \(Z\rightarrow \mu ^-\mu ^+\) and \(Z\rightarrow e^- e^+\)

To validate the accuracy of our code, we began by testing by detecting Z bosons using dilepton event data recorded using the ATLAS experiment at CERN from the ATLAS Open Data collection [ 28 ]. To confirm our code was functional, we would need to see a significant cluster of events around 91 GeV—the invariant mass of the Z boson [ 34 ]. Figure  18 (left) shows the results after initial cuts were made based on lepton flavour and charge, and, as expected, the resulting histogram had a high concentration of events with a reconstructed mass within the range of \(86-96\) GeV ( \(91\pm 5\) GeV). To gain a clearer view of the mass distribution, we removed lepton pairs that fell outside this mass range and increased the bin count, resulting in Fig.  18 (right). This observation of the Z peak in pairs of leptons with a combined invariant mass of approximately 91 GeV confirmed the reliability of the code.

Code validation through observing an event cluster at the invariant mass of the Z boson (91±5 GeV). The reconstructed invariant masses of both \(e^+e^-\) and \(\mu ^+\mu ^-\) pairs were plotted, both before (left) and after (right) an additional cut on dilepton invariant mass, to highlight the Z peak

6.5.2 Code validation using \(qq\rightarrow ZZ\rightarrow l^+l^-l^+l^-\)

To further validate the accuracy of our code, we instead analysed four-lepton events [ 18 ]. As before, to confirm that our code was reliable, we needed to see that a large fraction of the lepton pairs had an invariant mass within the range \(86-96\) GeV ( \(91 \pm 5\) GeV). We were expecting to plot a histogram with a similar shape to that of Fig.  18 . Figure  19 displays the reconstructed masses of same-flavour, opposite-sign lepton pairs in events with four leptons in the final state, separately for \(e^+e^-\) (left) and \(\mu ^+\mu ^-\) (right). Each plot has the Z peak shape we were expecting to see. Thus, we were able to conclude that our code was functional and that we had a solid foundation for further analysis.

ZZ production in the four-lepton channel. The invariant masses of \(e^+e^-\) and \(\mu ^+\mu ^-\) pairs are plotted separately for \(e^+e^-\) (left) and \(\mu ^+\mu ^-\) (right) pairs

6.5.3 Searching for the Higgs Boson: \(H\rightarrow \gamma \gamma\)

Having validated the reliability of the code, we explored the decay channel \(H\rightarrow \gamma \gamma\) [ 26 ]. Figure  20 depicts our results. Unfortunately, our analysis did not show a significant ’bump’ around the expected value of 125 GeV (the invariant mass of the Higgs boson [ 35 ]); therefore, we were unsuccessful in confirming its presence through the analysis of this decay channel.

A technique called ‘bump hunting’ was employed to detect the Higgs boson when conducting analysis of the \(H\rightarrow \gamma \gamma\) decay channel. We fitted a cubic function representing a data-driven estimate of background diphoton events (the red line) and normalised it (the grey line) to attempt to make the bump easier to see

6.5.4 Searching for the Higgs Boson: \(H\rightarrow W^+W^-\)

The last stage of our research involved conducting a non-resonant search to detect pairs of W bosons from dilepton data [ 28 ], shown in Fig.  21 (left). After using Monte Carlo simulations to estimate the background, displayed in Fig.  21 (middle), we subtracted this result from the data, to produce Fig.  21 (right). The presence of a Higgs signal in our resulting histogram suggested the presence of the Higgs boson. However, other factors must be considered before we can conclude that we detected the Higgs boson, e.g. we only accounted for one background process.

Each stage of performing a non-resonant search for \(H\rightarrow WW\) production. Selections are applied to dilepton data (left), \(qq\rightarrow WW\) backgrounds are estimated using MC simulation (middle), which were finally subtracted from data to produce our measurement of the Higgs signal (right)

6.6 Discussion

6.6.1 code validation using \(z\rightarrow \mu ^-\mu ^+\) and \(z\rightarrow e^- e^+\).

Our findings in Stage 1 demonstrate that our code was functional, since it accurately identified and reconstructed Z boson events from lepton pairs. As we hypothesised, there was a large concentration of events around 91 GeV, which indicated that the filters we used were effective in isolating events that originated from the decay of a Z boson. To make the mass distribution easier to see, we increased the number of bins and introduced a mass cut to filter events that were unlikely to have come from a Z boson decay. The results we achieved align well with research conducted by other groups in the past.

After conducting an initial round of analysis, we decided to implement a data caching mechanism in preparation for future stages. The selected events from the ROOT files we had downloaded and merged using the rootFile module were converted into an array and locally cached in a JSON file, exported using the pickle module. This means that if the code needed to be re-run, the JSON file could be loaded instead of going through the process of re-downloading and reanalysing the data, which would be time consuming.

Our results were promising, however, further refinements could be made to both the rootFile module prototype and our analysis code to make the data processing more efficient.

6.6.2 Code validation using \(qq\rightarrow ZZ\rightarrow l^+l^-l^+l^-\)

Stage two extended the validation process; however, rather than applying our code to events with two leptons, we applied it to events with four. This provided us with the opportunity to apply our code to a more complex scenario and check whether it was reliable enough to be used in further stages.

Recognising that by exploring events involving four leptons, we introduced the possibility that a single event could yield up to 4! valid pairs, we implemented a for loop that generated and analysed each of the 24 potential lepton pairs. Any pair that did not satisfy all the filters was discarded, and following this, only the pair that had a reconstructed mass closest to that of the Z boson (91 GeV), was included in the histogram. Although incorporating this step into our methodology extended the time taken for the data analysis, it proved worthwhile as it enabled us to identify a greater number of valid lepton pairs. Similarly to our approach during the initial stage, we employed the rootFile module to increase the efficiency of our analysis chain. We iteratively refined this module as we progressed to ensure it was fully functional by the time we began searching for the Higgs boson.

The results we achieved showed that our code was robust. There was a large peak of events around 91 GeV ( \(\pm 5\) GeV), which is what we hypothesised we would see. Figure  19 closely resembles Fig.  18 , which we had previously established was accurate. This strongly suggests that our code was reliable and could be used as the foundation for our data analysis in stages 3 and 4.

6.6.3 Searching for the Higgs Boson: \(H\rightarrow \gamma \gamma\)

Upon validating our code, the focus of our research shifted to the exploration of the \(H\rightarrow \gamma \gamma\) decay channel. As depicted in Fig.  20 , the data analysis did not yield the desired result: a ’bump’ around 125 GeV. Despite fitting a cubic function that represented a data-driven estimate of the background diphoton events to our histogram to make the \(H\rightarrow \gamma \gamma\) events easier to see, we were unsuccessful in confirming the presence of the Higgs boson.

Several factors could explain this. First, it is possible that our ‘tight’ requirements resulted in the discarding of many ‘good photons’. If we erred too far on the side of caution, trying to ensure that we did not involve events that did not originate from the \(H\rightarrow \gamma \gamma\) decay channel in our histogram, we may have missed valid photons originating from a Higgs decay. Not including these photons in our histogram may have resulted in the absence of the ’bump’ at 125 GeV. Secondly, we may not have analysed enough events to see the concentration of events around the invariant mass of the Higgs boson. \(H\rightarrow \gamma \gamma\) is a rare event and is often difficult to see over any background diphoton events. Additionally, the data-driven background fit to a cubic function may not have been robust enough to make the result visible on the histogram. It is possible that the use of a larger sample size may have resulted in the presence of the ‘bump’ at 125 GeV.

After realising that we had millions of events to analyse, which would take a significant amount of time, we decided to create a multithreading module. This module was made to speed up data processing by utilising threads to divide the workload into smaller sections that could run simultaneously. Instead of analysing each event one by one, multiple events were analysed at once, leading to a significant decrease in the time taken for the data processing.

Although we were unsuccessful in detecting the Higgs boson, we made improvements to the efficiency of our code through the development of the multithreading module, which proved especially beneficial for Stage 4 of our research.

6.6.4 Searching for the Higgs Boson: \(H\rightarrow W^+W^-\)

To search for the Higgs boson through analysis of the \(H\rightarrow W^+W^-\) decay channel, we conducted a non-resonant search. This was necessitated by the fact that the masses of the decay products were greater than the invariant mass of the Higgs boson, thus rendering the ‘bump hunting’ technique we had previously used not applicable. By subtracting the Monte Carlo simulated background from the real data, we estimated the Higgs signal, as depicted in Fig.  21 . Although this method did yield a Higgs signal, and therefore in theory confirmed the presence of the Higgs boson, it should be noted that our histogram only considered one source of background—diboson events. Had we considered multiple sources of background, the resultant histogram would have a much smaller Higgs signal, or possibly not one at all. Therefore, to confirm the presence of the Higgs boson, we would need to conduct further analysis on several other sources of background events, as well as the one we already considered, and re-examine the histogram.

6.7 Conclusions

Our research began with thorough code validation using \(Z\rightarrow \mu ^-\mu ^+\) and \(Z\rightarrow e^- e^+\) and \(qq\rightarrow ZZ\rightarrow l^+l^-l^+l^-\) events, which confirmed the reliability of our code in preparation for subsequent stages. While we did not yield results through exploring the \(H\rightarrow \gamma \gamma\) decay channel, we made significant improvements to our code, and developed a multithreading module to expedite data processing. After analysis of the \(H\rightarrow W^+W^-\) decay channel, we identified a Higgs signal. However, we could not confirm the presence of the Higgs, as we did not consider enough sources of background events to conclusively say that we detected it. Our iterative approach allowed for incremental improvements, and refining first drafts of code as we progressed through the stages streamlined the analysis process later. Our work demonstrated the benefits of such an approach and highlighted several areas for improvement, for example, the need for a larger sample size when exploring decay channels. Therefore, although our results did not confirm the presence of the Higgs boson, the code we utilised to explore our chosen decay channels provides us with a solid framework for future research endeavours.

7 Re-proving the existence of the Higgs Boson

By students of The Tiffin Girls’ School : Ayda Yazdani, Soniya Walke, Phoebe Lister.

7.1 Preface

In this section, students leverage the \(H\rightarrow \gamma \gamma\) and \(H\rightarrow WW\) s earch techniques developed in notebooks 6 and 7 to design a search in an additional decay channel: \(H\rightarrow ZZ\) . This is supported by skills in analysing ZZ production developed in notebook 5. Students also explore the idea of background fits, implementing an original fit to the continuum background to \(H\rightarrow \gamma \gamma\) to emphasises the Higgs mass peak.

7.2 Summary

The Higgs field plays a critical role in the Standard Model of particle physics. All massive elementary particles interact with the field through the Higgs mechanism and acquire mass, enabling them to form matter and give rise to the complex structures one can observe in the universe. The Higgs boson is the mediator the Higgs field.

The experimental method used to discover the Higgs boson involved the precise measurement of the properties of particles produced in proton–proton collisions at the Large Hadron Collider [ 2 ] using the ATLAS detector [ 1 ], and the use of statistical methods to identify the Higgs boson signal from the background events. We intended to apply these same skills on a smaller scale through our analysis of publicly available ATLAS Open Data [ 18 , 26 , 28 ] datasets from the ATLAS detector available on CERN’s website [ 8 ].

In our research, we investigated three decay channels of the Higgs boson: Higgs to two photons, Higgs to two W bosons and Higgs to two Z bosons. Executing cuts on the collection of ATLAS Open Data allowed us to target each of these channels individually and plot histograms to piece together evidence of the Higgs boson’s existence.

7.3 Introduction

The Standard Model of particle physics describes how the fundamental forces interact with particles. These elementary particles are classified into fermions, or ‘matter particles’, and bosons, known as ‘force carriers’, the latter which includes the Higgs boson.

Half a century after the Higgs mechanism was first proposed in the 1960s by the theoreticians Robert Brout, François Englert, and Peter Higgs, the existence of the Higgs boson was finally confirmed in 2012. Data was collected from the ATLAS [ 1 ] and CMS [ 29 ] detectors, at the Large Hadron Collider [ 2 ] at CERN to prove the existence of the particle. The measured data displayed a deviation from the expected backgrounds: in early ATLAS measurements, the invariant mass distribution of two photons produced in the diphoton channel showed a slight bump near 126 GeV (consistent with the Higgs boson’s hypothesised mass) with a significance of 2.2 standard deviations above the Standard Model (SM) background [ 36 ]. Later combinations with CMS results produced an observation of the Higgs boson at 125 GeV over 5 \(\sigma\) , above the threshold for discovery.

7.4 Research aims

This project aimed to re-prove the existence of the Higgs boson by exploring Higgs boson production in three of its most sensitive decay channels, each shown in Fig.  22 :

The \(H\rightarrow \gamma \gamma\) decay channel in the two-photon final state (2.8 \(\sigma\) local significance observed), looking for evidence of the Higgs boson in distributions of diphoton invariant mass, expecting a mass of 125 GeV;

The \(H\rightarrow WW\) decay channel in the two-lepton final state (1.4 \(\sigma\) local significance observed), using a non-resonant search technique;

The \(H\rightarrow ZZ\) decay channel in the four-lepton final state (2.1 \(\sigma\) local significance observed), by reconstructing the diboson invariant mass.

Feynman diagrams for two \(H\rightarrow \gamma \gamma\) decay modes (left) [ 37 ], one \(H\rightarrow WW\) decay mode (middle) [ 38 ], and one \(H\rightarrow ZZ\) decay mode (right)

7.5 Methods

We used the 13 TeV ATLAS Open Datasets [ 18 , 26 , 28 ], which provided us with data collected from real proton–proton collisions detected by ATLAS, in addition to simulated samples. These data files, presented in ROOT format [ 6 ], were analysed using a selection of Python libraries suited to our research: the uproot module for reading in data, the numpy library to carry out statistical analysis and the hist module from the larger matplotlib library for data visualisation and plotting histograms produced from our analyses.

7.6 Results

7.6.1 higgs to diphoton ( \(h\rightarrow \gamma \gamma\) ) channel.

In order to prove the existence of the Higgs boson through the Higgs to diphoton channel, we needed to plot the invariant mass of the two photons produced in selected event data, chosen after making suitable cuts to ensure that mainly events involving a two-photon system were included. The process, including the cuts involved, is enumerated below.

Diphoton invariant mass plots for datasets A-D

Loop through each data event in the TTree (a ROOT data storage format) from the ROOT file containing the collision data;

In each event, search for good-quality photons, which must:

Pass the diphoton trigger;

Pass “Tight” reconstruction requirements;

Have \(p_{\mathrm{T}}>25\) GeV;

Be in the ‘central’ region of ATLAS with \(|\eta | < 2.37\) , excluding the ‘transition region’ between ATLAS’s Inner Detector barrel and electromagnetic calorimeter endcap \(1.37 \le |\eta | \le 1.52\) ;

If the two photons are well-isolated, extract their four-momentum from the \(p_{\mathrm{T}}\) , \(\eta\) , \(\phi\) and energy, and store them in a TLorentzVector , a ROOT object emulated in the Jupyter notebook resources, which stores the energy and momentum of a particle as a four-vector. A function allowing us to extract invariant mass from a TLorentzVector was also provided;

Add the TLorentzVector s associated with the two photons together;

Calculate the invariant mass of the two-photon system;

Check each photon makes up a minimum fraction of the diphoton system invariant mass;

Fill a histogram with the invariant mass of the two-photon system.

We repeated this process for all the diphoton datasets, labelled A–D, provided by the ATLAS Open Data. We plotted separate histograms for each dataset, as shown in Fig.  23 .

Merged histogram for the diphoton invariant masses obtained from all events in A, B, C and 2,900,000 events of D (ran out of memory)

The datasets had varying numbers of events, with dataset D having the highest number (3,600,000 events), making the processing time longer for this dataset. Due to this, when merging the histograms for the different datasets to make the bump at the Higgs boson mass more visible, we were unable to include all events from dataset D, as we were limited by the processing power of our computers. The merged diphoton invariant mass histogram is shown in Fig.  24 .

To make the Higgs bump clearer to see, we produced a prediction of the background using a cubic function to fit the graph and plotted the data against this to make it stand out more. The background fits to each individual dataset are shown in Fig.  25 .

Fitted histogram for datasets A-D

7.6.2 Higgs to two W bosons ( \(H\rightarrow WW\) ) channel

Histogram of the transverse mass of the leptons produced in data as the decay products of the 2 W bosons (top left), histogram of the transverse masses of lepton from the simulated background events (top right), histogram of lepton transverse mass data with simulated backgrounds (above 2 graphs combined) (bottom left), and background-subtracted histogram showing the observed Higgs signal (bottom right)

To find a signal for the Higgs boson decaying to two W bosons, we used a non-resonant search technique to plot a histogram of the transverse mass of the decay products of the two W bosons. To perform this analysis, we first selected only data events with ‘good-quality’ leptons, using similar selections to the Higgs to diphoton channel.

For this channel, we then selected pairs of leptons with different flavours and opposite charges, to reduce contributions from background processes. We also placed requirements on the transverse momenta, with the leading and subleading leptons requiring \(p_{\mathrm{T}}>22\) GeV and \(p_{\mathrm{T}}>15\) GeV, respectively.

Then, we applied various event-level selection criteria, including the magnitude of the missing transverse energy (MET) and the angle \(\phi\) between the MET and the dilepton system, to ensure that the selected events were consistent with the \(H\rightarrow WW\) signal and that events that may have arisen from background processes were rejected.

Next, we used the samples of \(qq\rightarrow WW\) Monte Carlo simulations provided by the ATLAS Open Data to model the expected background contributions, scaled to match the luminosity of experimental data used. The same selections that were applied to the data were also applied to the simulation, and the transverse mass of the dilepton system in the simulated events was also plotted in a histogram. The final step was to subtract the background from the data, and this produced our Higgs signal.

Each of the steps described above is shown in Fig.  26 .

7.6.3 Higgs to two Z bosons ( \(H\rightarrow ZZ\) ) channel

Higgs to ZZ channel histogram

To investigate the Higgs to ZZ decay channel in the four-lepton final state, we modified the code used in Sect.  6.5.2 to instead plot the invariant mass of the Z boson. Instead of using datasets with events in the two-lepton final state, we used data events from the datasets with events in the four-lepton final state.

Slightly different selection criteria were applied to the datasets, allowing only events with two pairs of same-flavour opposite sign leptons to target the decay of a Z boson. These pairs were used to reconstruct two pairs of TLorentzVector s, corresponding to the leading and trailing Z boson. The two reconstructed Z bosons were then combined to reconstruct the four-momentum of the ZZ system, which was used to plot the histogram of the system’s invariant mass. Figure  27 shows the results of this method using the events from Dataset D of the ATLAS Open Data four-lepton final-state data collection. There is a clear peak at 125 GeV, which is the accepted value for the invariant mass of the Higgs boson. This is evidence that the Higgs boson exists and was produced during several of the events in the dataset used.

7.7 Discussion

After examining all our histograms, it was clear that there was a pronounced bump at 125 GeV for the \(H\rightarrow \gamma \gamma\) and \(H\rightarrow ZZ\) channels, the expected invariant mass we would expect from the Higgs boson, in addition to a clear Higgs signal in the \(H\rightarrow WW\) . Since we conducted multiple investigations through different decay channels, it solidified our evidence for observing the Higgs boson.

We ensured that we worked well as a team and divided up sections to focus our analysis on, and met regularly to share our findings and create a plan of our next steps. It also meant we could work through coding challenges together, as we were able to share similar experiences and collaborate to find a solution. We also watched some lectures and videos to enable us to understand the relevant physics before diving deep into our research, so that we could get the most out of the project.

7.8 Acknowledgements

We would like to thank the Institute for Research in Schools, Rutherford Appleton Laboratory, and the University of Oxford for providing us with this enriching opportunity. Moreover, thank you to our teacher, Mr Carpenter, and coordinator, Dr Richard Phillips, for supervising and supporting us. In addition, a special thank you to Professor Alan Barr and his Ph.D. students for their guidance and for inspiring us to pursue this project.

8 Searching for the Higgs boson through its decay into a muon-antimuon pair

By students of King Edward VI Camp Hill School for Boys : Rohan Desai, Yijun Chen, Amogh Shetty, Ishaan Dubey.

8.1 Preface

In this section, students use the ATLAS Open Data and the skills developed in all seven notebooks to design an original search for a rare Higgs boson decay: \(H\rightarrow \mu \mu\) . The students also implement several ideas in statistics, such as kernel density estimation to perform a continuous fit to a histogram, signal significance, and p-values.

8.2 Summary

Using Atlas Open Data [ 28 ], we searched for the predicted decay of the Standard Model (SM) Higgs boson into a muon-antimuon pair. The Large Hadron Collider (LHC) [ 2 ] provided us with data at \(\sqrt{s}\) = 13 TeV from proton–proton collisions. By imposing selection criteria, we isolated events that are most likely to exhibit the characteristics of our desired decay. We reconstructed the masses of these events using the TLorentzVector class, provided in the Jupyter [ 7 ] notebooks, to calculate our invariant dimuon mass, \(m_{\mu \mu }\) , which we used to populate a histogram. We used Monte Carlo (MC) files from the ATLAS Open Data to simulate the backgrounds of this decay, which we then subtracted from the real data to isolate the signal. After processing 9.4 million MC events and 12.2 million data events, a peak was revealed within the range for the mass of a Higgs boson, Footnote 1 \(m_{\mathrm{H}}\) , at 125.66 GeV. The observed significance for a Higgs boson at \(m_{\mathrm{H}}=125.38\)  GeV was calculated to be 1.169 \(\sigma\) . While not statistically significant to the level of observation, this result supports the possibility of the decay of a Higgs boson to second-generation fermions.

8.3 Introduction

All particles are proposed to be excitations of fields—the Higgs boson is an elementary particle in the Standard Model and an excitation of the Higgs field, which gives mass to elementary particles. The Higgs field is a scalar field; therefore, its associated boson is scalar and has a spin of zero. The addition of this field allows spontaneous symmetry breaking of the electroweak interaction, giving mass to many particles via the Higgs Mechanism.

The Higgs field can be analogised to crossing an infinite, flat field of snow. The following scenarios are possible:

Skiing across the top—this is analogous to a high-energy particle not interacting with the Higgs field. It does not sink into the ‘snow’ as it is travelling at the speed of light and therefore has no mass.

Walking in snowshoes—they will sink into the ‘snow’ as they travel slower than before. This is like a particle with some mass as this person somewhat interacts with the field.

Walking regularly—this person will sink deeply into the field, as they are travelling very slowly and with little energy. This represents a particle with greater mass that interacts strongly with the field.

Just as the snowfield is made up of tiny, individual snowflakes, the Higgs field gives rise to many Higgs boson excitations, which give mass to elementary particles.

The Higgs boson was discovered in 2012 by the ATLAS [ 1 ] and CMS [ 29 ] experiments at the Large Hadron Collider (LHC) [ 2 ] at 5.9 \(\sigma\) significance [ 41 ]. It is measured to have an invariant mass of \(125.38\pm 0.14\hbox { GeV}\) [ 39 ], and is found to be consistent with the predicted properties for the Higgs boson by Peter Higgs et al. in 1964 [ 42 ]: Even (positive) parity, no electric charge, no colour charge, zero spin, and zero strong force interaction. Even the Higgs branching ratios have agreed with those predicted. The first evidence of fermion interactions with the Higgs field was through Higgs decay to tau particles, which was observed in the combination of ATLAS and CMS results performed at the end of Run 1 at the LHC, and later remeasured at a higher significance [ 43 ].

The Higgs boson is very unstable, with a lifetime of \(1.6\times 10^{-22}\) seconds [ 44 ], which means it decays almost immediately, making it difficult to find. By measuring decay rates to different particles, the predicted mechanism by which they acquire mass can be tested. Measurements performed so far have focused on Higgs boson interactions with the most massive particles, such as the W and Z bosons, and only with particles from the most massive generation, the top and bottom quarks and the tau lepton. The interaction of the Higgs boson with lighter particles, such as muons, has so far not been observed. Measuring the full spread of Higgs boson interactions is critical to test if the Higgs mechanism can explain the full range of particle masses.

The Standard Model predicts several rare Higgs boson decay channels which have not yet been observed. Among these are decays to second-generation leptons and quarks, e.g. \(H\rightarrow \mu \mu\) , and \(H\rightarrow Z\gamma\) . The focus of this project is on one of the rarest decays: the Higgs boson into a dimuon pair ( \(H\rightarrow \mu \mu\) ). The expected branching fraction for the decay of the Higgs boson into a pair of muons at \(m_{\mathrm{H}}=125.38\)  GeV is \(B(H\rightarrow \mu \mu ) = 2.18 \times 10^{-4}\) [ 45 ]. Figure  28 displays this statistic more visually. Other more prevalent decays have significantly higher branching fractions, making them easier to detect.

The branching ratios of a 125.38 GeV Higgs (left), created using data from [ 46 ], the branching ratios of the Higgs with respect to the mass of the Higgs (right), from [ 47 ]

Only one in five thousand Higgs bosons is predicted to decay to muons. And, like a needle in a mountain of needles, for every predicted decay of a Higgs boson to muons at the LHC, there are a thousand pairs of muons that mimic our desired signal [ 45 ]. This background from other particles makes isolating the Higgs boson decay to muons extremely difficult. Therefore, the efficacy of event selection and simulation is paramount. The \(H\rightarrow \mu \mu\) decay offers the best opportunity to measure the Higgs interaction with second-generation fermions at the LHC, providing new insights into the origin of mass for different generations.

8.4 Methods

The ATLAS experiment [ 1 ] is located at the LHC [ 2 ], which collides protons at high speeds and uses a set of complex detectors to measure the outcome. Data from these collisions are organised into ’events’, some of which have been made available to the public. We used files from the ’ATLAS Open Data’ [ 28 ] to conduct our investigation. The ‘bump hunt’ method that we employed is commonly used for measurements at CERN - using the law of conservation of energy, the Higgs boson can be found by reconstructing the invariant masses of its decay products ( \(m_{\mu \mu }\) ). After selecting appropriate events using the selection criteria in Table 1 , each dimuon invariant mass is added to a histogram. This process was repeated millions of times, so the final histogram should have a higher frequency around the mass of a Higgs boson (125.38 GeV)—a ‘bump’ suggesting existence of the \(H\rightarrow \mu \mu\) decay. The same process can then be carried out using simulated data for the backgrounds to this decay, and this background is subtracted from the data to enhance the visibility of our bump. A statistical analysis is then performed to quantify this evidence.

8.4.1 Event selection

To find this extremely rare decay, we used selection criteria: a set of filters applied to the data to distinguish the relevant signal from background data. After extensive research [ 48 , 49 ], we produced the selection criteria shown in Table 1 to identify and isolate the specific events where this dimuon decay has occurred. The same selection criteria were used to filter the real and simulated events.

8.4.2 Simulated events

We simulated the collisions of particles in the LHC using files from the 13 TeV ATLAS Open Data set [ 28 ]. Monte Carlo (MC) simulation files are used to simulate the behaviour of subatomic particles produced by the LHC. We used electroweak diboson and Higgs MC files from the Open Data, which we believed would mimic the expected background of the \(H\rightarrow \mu \mu\) decay. The simulated events were filtered through the same selection criteria as in Table 1 .

8.4.3 Invariant mass reconstruction and plotting

Using the selection criteria in Table 1 , we found the data events most likely to exhibit the characteristics of the \(H\rightarrow \mu \mu\) decay and background MC events that mimic \(H\rightarrow \mu \mu\) . Using the implementation of ROOT’s [ 6 ] TLorentzVector class in the training notebooks, we used the transverse momentum, pseudorapidity, azimuthal angle and energy of these events to reconstruct the four-momentum of our desired decay. The notebook implementation of the ROOT SetPtEtaPhiE and M functions allowed us to easily calculate \(m_{\mu \mu }\) for each event and add it to a histogram, forming the red histogram of ’real data’ in Fig.  29 . We used ‘kernel density estimation’ (KDE) to transform our discrete histogram into a continuous line, improving our ability to find patterns in the distribution. The Gaussian distribution was our “kernel”, and we used Scott’s rule to calculate the bandwidth of the kernel, giving us an appropriate resolution for the data.

The line formed after using a Gaussian KDE to transform the histogram of the real dimuon masses, in red. The green line shows the same for Monte Carlo simulated events. The y -axis displays frequency, with the simulated graph being scaled up to have the same proportions as the real data

The same process was repeated for the simulated events; this formed the red and green lines for real and simulated data, respectively, in Fig.  29 . As there was a difference in the number of real and simulated events produced, we scaled the simulated graph up to have the same normalisation as the distribution of real events. This step helped us compare the simulated background data to real data, allowing us to identify events consistent with the expected behaviour of \(H\rightarrow \mu \mu\) decay and remove events that are the result of other background processes. Any difference in the two lines in Fig.  29 suggests a deviation from the expected behaviour, providing evidence for the \(H\rightarrow \mu \mu\) decay.

8.5 Results

Graph of weighted data events with simulated backgrounds subtracted

After applying the event selections in Table 1 to the reconstructed data and simulation, as shown in Fig.  29 , we were able to create Fig.  30 by subtracting the simulated background from the data, therefore isolating the signal. In total, we processed 12.2 million real data events and 9.4 million MC events, with the latter scaled up to have equal luminosity as the data. Our selection criteria were able to filter these background events to approximately 35% of their original number. There is a significant spike at 125.5 GeV, which lies in the 1 \(\sigma\) confidence interval for the Higgs mass (125.38±0.15 GeV). Due to the rarity of this decay, we were unable to isolate all the signal from the background, leaving us with bumps/ troughs at m \(\simeq\) 121 GeV and 128 GeV, respectively. The bump in MC data in Fig.  29 between 128 and 129 GeV appears to show the presence of another background decay that survived our selection criteria. This highlights the challenges in isolating rare events from background noise. Ideally, we would investigate this decay in order to possibly isolate the signal data further, which would produce a clearer spike.

Plot of observed local p-values of results at a range of test masses for the Higgs boson. Dotted lines show the corresponding 1 \(\sigma\) and 2 \(\sigma\) values

From our results, we conducted a statistical analysis to find the p values ( p ) and the sigma values ( \(\sigma\) ), testing the Higgs boson mass hypothesis over an interval. We used the Gaussian function to calculate the error in our methods to calculate the data and background for each mass, then we used the data and background values from Fig.  29 to calculate the \(\sigma\) and p value. This process was repeated between 122 and 128 GeV, creating Fig.  31 which displays a p value against the tested Higgs mass. At \(m_{\mathrm{H}}=125.38\)  GeV, the accepted value for the Higgs mass, our observed \(\sigma =1.169\) ( \(p=0.121\) ). However, we found at \(m_{\mathrm{H}}=\) 125.66 GeV, \(\sigma = 1.224\) ( \(p=0.111\) ). The periodic fluctuation in our graph was a point of interest, something we hope to understand and improve on in the future. Further data could be processed to give a larger sigma value, as well as exploring the questions outlined in Sect.  7.5 .

8.6 Discussion

From the results of our research, we have found evidence in support of the existence of a Higgs decay into a muon-antimuon pair. Figures  29 and 30 show that there was an excess of events around 125 GeV, which is consistent with the expected signal from the Higgs boson decay. More precise results would be needed to confirm this as a discovery. The results from different Higgs decay modes could be combined to improve the precision of the measurements and provide a more complete understanding of the Higgs boson’s properties.

Another improvement to this project would be to divide the decay into its production categories, a method that previous searches have used [ 48 , 50 ]. Specifically, there are four exclusive categories of Higgs boson production: Gluon-gluon fusion (ggf), association with vector boson (VH), vector boson fusion (VBF), and association with the top quark and antiquark pair (ttH). VBF has shown to be dominant out of the four; its events could be investigated exclusively to obtain better results. Our next steps would be to apply specific selection criteria for each of these categories to refine our search for the dimuon pair, likely resulting in a bigger peak. However, it would be difficult to separate events to that degree with our dataset, as that level of refining would leave each category with a low number of events, resulting in larger uncertainty and less reliable results.

Although the ATLAS Open Data were very useful in providing us with the necessary information and resources to develop our skills, we found that some of the data we would have liked were not provided. For our decay ( \(H\rightarrow \mu \mu\) ), we would have liked to have been provided with a simulation of the \(H\rightarrow \mu \mu\) signal and certain specific backgrounds, but these were not available. Instead, we had to compromise the accuracy of our results by combining similar processes. An alternative would be to simulate this background ourselves, using open-source software such as MadGraph or Delphes. However, MadGraph does not model particle interactions with the ATLAS detector, and while Delphes does, it is not approved by ATLAS. Therefore, there was no way to accurately simulate the background ourselves, highlights a limitation of the ATLAS Open Data to motivated student researchers. Although some alternatives would have given better simulation results than ours, none would be comparable to those used at CERN. Furthermore, using machine learning algorithms such as XGBoost to identify b -tagged jets could more accurately isolate the signal data.

When evaluating our project, we found that we had initially overlooked the process of converting the histogram into a continuous curve. We used a simple kernel density estimation and found a width using Scott’s rule, but after further research we understood that we should have controlled this process further. One ATLAS paper stated “The width of the Gaussian component of the double-sided Crystal Ball function varies between 2.6 and 3.2 GeV depending on the category” [ 50 ]. A width between 2.6 and 3.2GeV would have given a less sensitive but more appropriate resolution for this investigation. Our function also took into account the edges of the histogram, as seen in 29 , where the graphs fall significantly on each side. If we were to repeat this process, a more appropriate function would be used.

With our observed \(\sigma =1.224\) , this project supports the prediction of the \(H\rightarrow \mu \mu\) decay and provides evidence for the decay of the Higgs boson to second-generation fermions. Although the ATLAS Open Data was extensive, an extension to the data provided would have resulted in a better result.

8.7 Acknowledgements

We would like to thank the Institute of Research in Schools, Rutherford Appleton Laboratory, and Oxford University for providing the data and learning modules to develop our ideas and conduct research, as well as the opportunity to present our findings at the IRIS Student Research Conference in London. We would also like to thank William Murray from Rutherford Appleton Laboratory for helping us to conduct a statistical analysis to find our sigma values. Finally, we would like to thank our school teacher Daniel Redshaw for supervising the project.

9 Applications of the XGBoost machine learning algorithm in particle physics

By students of King Edward VI Camp Hill School for Boys : Shizhe Liu, Sasan Hapuarachchi, Pruthvi Shrikaanth, William Shi, Joel Regi.

9.1 Preface

In this section, students utilise the ATLAS Open Data and skills developed in all seven notebooks to explore the potential of machine learning in particle physics analyses. Students compare traditional ‘cut and count’ methods to the output of the XGBoost classifier for different Standard Model and Beyond Standard Model processes. Students evaluate the performance of the machine learning algorithm using ROC curves and the Approximate Mean Significance metric.

9.2 Abstract

The rise in technological developments in artificial intelligence has opened up new avenues of exploration at the intersection of machine learning (ML) and particle physics. We evaluated the potential of the XGBoost (ML) algorithm, a powerful gradient-boosted decision tree classification algorithm, to streamline the process of identifying rare particle decays. We achieved this by comparing the performance of XGBoost in four different classification problems in particle physics with the performance of existing classification methods, such as the application of strict cuts. We found that while in some cases the algorithm provided near-perfect prediction results, the algorithm was overly rigorous in other cases, leading to large numbers of signal events being dismissed as background events by the algorithm.

9.3 Introduction

9.3.1 motivation.

Currently, the process of searching for rare particle decays presents a significant challenge for particle physicists, as these decays can only be found in a tiny proportion of the millions of events picked up by the sensors in particle detectors. The emergence of the XGBoost, a powerful classification algorithm, has the potential to aid particle physicists in resolving this challenge, as it may provide a better alternative to existing methods in identifying the events that contain elusive particle decays.

However, there are limited studies on the performances of the XGBoost algorithm in particle physics classifications. Therefore, we aim to address this gap by identifying the specific areas within particle physics where XGBoost excels and to compare its performance with existing methods, such as applying stringent cuts. We evaluated the algorithm’s classification performance in four different event types:

Higgs boson events;

Supersymmetry events;

Beyond standard model \(Z^{'}\) events;

Kaluza–Klein graviton events.

9.3.2 XGBoost classification algorithm

As advances in artificial intelligence continue to be made, increasingly powerful machine learning algorithms are being developed at a rapid pace, and the XGBoost algorithm is an example of a classification algorithm that has arisen as a result of this technological revolution. It uses gradient boosted decision trees to provide accurate classification results. ‘Boosting’ is a technique where new models correct errors made by previous ones, and are added one by one until no further improvements occur. ‘Gradient boosting’ allows models to predict the errors made by the previous models, to help provide an accurate result. The main advantages of XGBoost are its excellent speed and accuracy, due to its ability to discern subtle patterns in the data, in providing accurate predictions that may prove invaluable when performing particle physics classifications [ 51 ].

9.3.3 AMS metric

We evaluated the performance of the XGBoost classifier algorithm in carrying out the different categories of classification using the Approximate Median Significance (AMS) metric. When providing the true positive and false positive rates of the classification, the AMS metric uses the Wilks Theorem to compare the probabilities of observing the signal+background hypothesis and the background only hypothesis, to return an overall figure which represents the performance of the classifier. Therefore, AMS provided us with a standardised way of assessing the XGBoost algorithm in the different applications tested [ 52 ].

9.4 Results

9.4.1 higgs boson event classification.

Prior advancements in particle physics have revealed that it is possible to represent every particle as a wave in quantum fields. One such field is the Higgs field, which suggests that there would be a particle associated with this field, the Higgs boson [ 53 ]. We wrote a programme that trained the XGBoost algorithm to classify Higgs boson events using ATLAS Open Data samples containing over 400,000 events, including a mixture of simulated signal and background events [ 54 ], and tested it with another set of 400,000 events. All the test events that the XGBoost classifier had classified as signal events were, indeed, a real signal event. This yields an AMS score of infinity, which suggests that the XGBoost algorithm performed very well in classifying Higgs boson events. This can be reinforced by observing the ROC curve, which compares the true positive rate (TPR) against the false positive rate. As evident in Fig.  32 , the TPR initially increases rapidly, further suggesting that the XGBoost is accurate.

Higgs boson classification ROC curve

9.4.2 Beyond standard model \(Z^{'}\) event classification

The second type of particle we looked at is the \(Z^{'}\) boson hypothesised by the Topcolor model—a model for electroweak symmetry breaking, in which a top anti-top pair forms a composite Higgs boson. In this model, a \(Z^{'}\) boson is predicted to exist, which we decided to investigate via the XGBoost classifier.

The invariant mass of \(Z^{'}\) events classified using strict cuts (left), and \(Z^{'}\) events classified using the XGBoost algorithm (right)

To find these particles, we looked at the decay channel; in this case, the \(Z^{'}\) decays into a top-antitop pair in events with a single charged lepton, large-radius (large-R) jets and missing transverse momentum [ 17 ]. The lepton must have a transverse momentum > 30 GeV, missing transverse energy > 20 GeV, a small radius (small-R) jet close to the lepton, a large-R jet passing the top tagging requirements (mass > 100 GeV, N-subjettiness ratio < 0.75), etc.

Finally, we plotted the histograms of the \(Z^{'}\) invariant masses from the cut-based analysis (Fig.  33 (left)) and the results of the XGBoost classifier (Fig.  33 (right)).

9.4.3 Kaluza–Klein graviton event classification

The third type of particle we looked at is a Kaluza–Klein graviton hypothesised in the Randall–Sundrum model [ 55 ], a model for gravity in which gravity propagates through warped extra dimensions. Similarly to atoms having excited states or low energy states, particles can have corresponding Kaluza-Klein states where the particle has extra mass in other dimensions [ 56 ].

To find the Kaluza–Klein graviton, we searched for the particles into which it decays, in this case a \(\gamma \gamma\) pair. To find this particle, we performed a bump hunt in photons with transverse energies over 20 GeV [ 55 ]. To do this, we made the following cuts to the ATLAS Open Data set [ 26 ]: the event must have two photons, it must activate the photon trigger, and both photons must have a transverse energy greater than 20 GeV. Once we obtained our data points, we subtracted any that could have been produced by a \(H\rightarrow \gamma \gamma\) decay and plotted a graph of invariant mass against frequency using a fitting function.

Kaluza–Klein graviton events classified using restrictive cuts (left), and Kaluza–Klein graviton events classified using the XGBoost algorithm (right)

XGBoost classifiers were trained on ATLAS Open Data samples containing information (e.g. number of leptons, transverse mass, jet) to identify events which had good-quality photons and isolated photons, respectively—this problem required two classifiers. The models were then used in parallel to identify events which contained the decay of the Kaluza–Klein graviton, which typically had both good photons and photon isolation. The events classified by XGBoost were then plotted on a histogram shown in Fig.  34 (right) and compared with the original cuts-based histogram shown in Fig.  34 (left) to see the performance of the classification. The AMS values were 866.8 for the good photon classification and 866.7 for the photon isolation classification.

9.4.4 Supersymmetric event classification

‘Supersymmetry’ (SUSY) is the hypothesis that every fermion has a partner boson with different spin properties, where fermions have half-integer spin values and bosons have integer spin values [ 57 ]. We can search for supersymmetric particles by examining pairs of particles created from collisions in the LHC. To do this, we used Python to examine the ATLAS Open Data [ 28 ] and make ’cuts’ on it. These cuts filter out data that we do not need, leaving us with a subset of data that is much more useful for examining supersymmetric events.

SUSY events general classification using restrictive cuts (left), and SUSY events general classification using the XGBoost algorithm (right)

To carry out this investigation, we first used uproot to load the ATLAS Open Data and initialised our histograms to be plotted later. After this, we extracted all the information we needed to make our cuts from the data and stored it. Then, we set up variables to store the four-momentum of particles by creating four-vectors of their kinematics. A series of cuts were made, starting with selecting only collisions between electrons and muons in pairs of the same type and opposite charge. We then selected events where each particle had a minimum momentum, which we decided by following recommendations from the ATLAS experiment at CERN. Following this, we calculated the momentum of leading and trailing leptons, and manipulated their four-vectors to find the dilepton invariant mass. We then made further cuts based on the invariant mass and the accuracy of detected jets. Next, we sorted these leptons into categories based on the magnitude of their invariant mass and the event’s MT2 variable [ 58 ], which is related to the transverse mass of unseen particles. Finally, we plotted the distributions of the dilepton invariant mass with general (least strict), loose and tight requirements on dilepton invariant mass and MT2 values.

SUSY events loose classification using restrictive cuts (left), and SUSY events loose classification using the XGBoost algorithm (right)

After obtaining these results, we stored them in a.csv file. We then used the XGBoost machine learning algorithm, training it on half of our data, and testing it on the other half to see how well it matched the results of the cut-based analysis. To find the precision of our ML algorithm, rather than simply calculate the proportion of predictions that were ’correct’ or within an acceptable range, we used the AMS metric, defining a function that implements the AMS metric and then calculates it for each category (general, loose, tight). The graphs plotted by the XGBoost algorithm are given in Figs.  35 , 36 and 37 . The algorithm produced graphs, and AMS values \(\sim\) 1.689 (loose) and \(\sim\) 1.192 (tight), while the general category had an AMS value of \(\sim\) 603.226. This may have been a result of using a smaller dataset as this would have resulted in a weaker model. The comparison of the graphs shows us that our loose events classification predicted by the XGBoost algorithm shown in Fig.  36 (right) is most similar to the loose events classification made using cuts shown in Fig.  36 (left), sharing a shape with the tight classification graphs shown in Fig.  37 . Hence, we found that loose cut requirements were the best for building an accurate model to detect supersymmetric particles, although they may lead to more false positives than desirable when compared to tight requirements.

SUSY events tight classification using restrictive cuts (left), and SUSY events tight classification using the XGBoost algorithm (right)

9.5 Discussion

From BSM particles to delving into supersymmetry, we have explored the performance of the XGBoost machine learning algorithm across a wide range of cutting-edge classification problems in particle physics.

Our results revealed significant variations in the performance of the XGBoost algorithm across our tested range of particle physics classification problems. In some areas, such as the Higgs boson classification, the XGBoost gave perfect or near-perfect prediction results. However, in other cases, it was clear that the XGBoost displayed excessive rigidity, leading substantial portions of the signal data to be excluded and dismissed as background data, resulting in lacklustre distributions, as seen in our exploration of SUSY.

The XGBoost is a supervised learning algorithm, and so relies on labelled datasets. Therefore, the algorithm works best in classifications where the selection criteria are well defined, as it allows accurately labelled training datasets to be generated. In these cases, researchers may not feel the full benefit of the XGBoost, as the algorithm will only ever (at best) replicate a prior cut-based analysis. However, this is an issue that we hope to address in our future work by exploring the potential of deep learning models to identify the optimal selection criteria for particle decay classifications which have very few selection criteria identified so far.

10 Conclusion

In this article, a variety of original research projects performed by UK secondary school students using the ATLAS Open Data and the repository of training resources developed by the authors have been presented. Such student research output shows that secondary school students are capable of meaningfully engaging with public releases of LHC data, presupposing no prior knowledge or experience. With sufficient time, training and support, it has been demonstrated that it is possible for high school students to interact with the data presented in the same format and using the same analysis techniques as physics researchers. Additionally, it has been shown that, with correctly structured training, students can produce entirely original works of research, and often independently arrive at questions and ideas that exist at the cutting-edge of particle physics research. Key to such successes is the structure of the training materials; they must presuppose no prior knowledge and present new information in a step-wise manner (preferably in a variety of formats), coding examples should be thoroughly commented with interleaved exercises to consolidate learning, hints and solutions should be available throughout to prevent frustration due to students becoming ’stuck’. Crucially, off-ramps from the training should be provided at each level, so individual teachers can tailor the project to the particular group of students, and the time and resources available. In conclusion, the student research presented in this article makes a strong case for the value of the public release of LHC data, and for the ongoing support for the ATLAS Open Data project.

Data Availability Atatement

The data and simulation that support the findings of this study are openly available in the ATLAS Open Data repository, at http://doi.org/10.7483/OPENDATA.ATLAS.GQ1W.I9VI , http://doi.org/10.7483/OPENDATA.ATLAS.B5BJ.3SGS , http://doi.org/10.7483/OPENDATA.ATLAS.2Y1T.TLGL and http://doi.org/10.7483/OPENDATA.ATLAS.FRWJ.4ZQU . The manuscript has associated data in a data repository.

At the time of writing, it is accepted that \(m_{\mathrm{H}}\) = 125.38±0.14 GeV [ 39 ]. Recent research has shown that \(m_{\mathrm{H}}\) = 125.11±0.11 GeV may be possible [ 40 ].

ATLAS Collaboration: The ATLAS Experiment at the CERN Large Hadron Collider: A Description of the Detector Configuration for Run 3 (2023) arXiv:2305.16623 [physics.ins-det]

L. Evans, P. Bryant, LHC Machine. JINST 3 (S08001) (2008) https://doi.org/10.1088/1748-0221/3/08/S08001

Review of the 13 TeV ATLAS Open Data release. Technical report, CERN, Geneva (2020). https://cds.cern.ch/record/2707171

Review of ATLAS Open Data 8 TeV datasets, tools and activities. Technical report, CERN, Geneva (2018). https://cds.cern.ch/record/2624572

ATLAS Data Access Policy. Technical report, CERN, Geneva (2015). https://cds.cern.ch/record/2002139

R. Brun, F. Rademakers, Root: an object oriented data analysis framework. Nucl. Instrum. Methods Phys. Res. Sect. A 389 (1), 81–86 (1997). https://doi.org/10.1016/S0168-9002(97)00048-X

Article   ADS   Google Scholar  

B.E. Granger, F. Pérez, Jupyter: thinking and storytelling with code and data. Comput. Sci. Eng. 23 (2), 7–14 (2021). https://doi.org/10.1109/MCSE.2021.3059263

Article   Google Scholar  

ATLAS Open Data and Tools for Education. https://opendata.atlas.cern/

notebooks-collection-opendata. https://github.com/eimearconroy/notebooks-collection-opendata/tree/master

opensource.com: What is Python? https://opensource.com/resources/python . Accessed: 26-02-2024. (No date provided)

A. Corbo, What is Python? https://builtin.com/software-engineering-perspectives/python . Accessed: 17-04-2024 (2022)

Lumen Learning: Projectile Motion. https://courses.lumenlearning.com/suny-physics/chapter/3-4-projectile-motion/ . Accessed: 17-04-2024 (No date provided)

CERN: The Large Hadron Collider. https://home.cern/science/accelerators/large-hadron-collider . Accessed: 17-04-2024 (2024)

The ATLAS Experiment: The ATLAS Experiment. https://atlas.cern/about . Accessed: 17-04-2024 (2024)

CERN: ATLAS. https://home.cern/science/experiments/atlas#:~:text=ATLAS%20is%20one%20of%20two,could%20make%20up%20dark%20matter . Accessed: 17-04-2024 (2024)

CERN: The Standard Model. https://home.cern/science/physics/standard-model . Accessed: 17-04-2024 (2024)

ATLAS Collaboration: ATLAS 13 TeV samples collection one large-R jet and one lepton, for 2020 Open Data release. CERN Open Data Portal (2020). https://doi.org/10.7483/OPENDATA.ATLAS.FRWJ.4ZQU

ATLAS Collaboration: ATLAS 13 TeV samples collection at least four leptons (electron or muon), for 2020 Open Data release. CERN Open Data Portal (2020). https://doi.org/10.7483/OPENDATA.ATLAS.2Y1T.TLGL

P. Mouche, Overall view of the LHC. General Photo (2014). https://cds.cern.ch/record/1708847

J. Pequenao, P. Schaffner, How ATLAS detects particles: diagram of particle paths in the detector (2013). https://cds.cern.ch/record/1505342 .

\(W\) and \(Z\) bosons: Wikipedia contributors. Wikipedia, The Free Encyclopedia. Accessed: 17-04-2024 (2023). https://en.wikipedia.org/w/index.php?title=W_and_Z_bosons &oldid=1173699311 .

Monte Carlo method: Wikipedia contributors. Wikipedia, The Free Encyclopedia. Accessed: 17-04-2024 (2023). https://en.wikipedia.org/w/index.php?title=Monte_Carlo_method &oldid=1175789321

Institute for Research in Schools: Overview—Projects. https://researchinschools.org/projects/ . Accessed: 17-04-2024 (2023)

B. Greene, How the Higgs Boson Was Found. https://www.smithsonianmag.com/science-nature/how-the-higgs-boson-was-found-4723520/ . Accessed: 17-04-2024 (2013)

Particle Data Group, Review of particle physics. PTEP 2022:083–01 (2022). https://doi.org/10.1093/ptep/ptac097

ATLAS Collaboration: ATLAS 13 TeV samples collection Gamma-Gamma, for 2020 Open Data release. CERN Open Data Portal (2020). https://doi.org/10.7483/OPENDATA.ATLAS.B5BJ.3SGS

CERN: The Higgs boson. https://home.cern/science/physics/higgs-boson . Accessed: 17-04-2024 (2024)

ATLAS Collaboration: ATLAS 13 TeV samples collection with at least two leptons (electron or muon), for 2020 Open Data release. CERN Open Data Portal (2020). https://doi.org/10.7483/OPENDATA.ATLAS.GQ1W.I9VI

CMS Collaboration: The CMS experiment at the CERN LHC. JINST 3 (S08004) (2008). https://doi.org/10.1088/1748-0221/3/08/S08004

CERN: The Large Hadron Collider. https://home.web.cern.ch/science/accelerators/large-hadron-collider . Accessed: 17-04-2024 (2023)

N. Wolchover, L.R.-I.S. Velasco, A New Map of All the Particles and Forces. https://www.quantamagazine.org/a-new-map-of-the-standard-model-of-particle-physics-20201022/ . Accessed: 17-04-2024 (2020)

CERN: What’s so special about the Higgs boson? https://home.cern/science/physics/higgs-boson/what . Accessed: 17-04-2024 (2023)

ATLAS Collaboration: Mass / Invariant mass. https://atlas.cern/glossary/mass . Accessed: 17-04-2024 (2024)

E. Conroy, Intro to finding Z & Higgs Bosons (University of Oxford, Oxford, 2021)

Google Scholar  

CMS Collaboration, Observation of a new boson at a mass of 125 GeV with the CMS experiment at the LHC. Phys. Lett. B 716 (1), 30–61 (2012). https://doi.org/10.1016/j.physletb.2012.08.021

ATLAS Collaboration: Combined search for the Standard Model Higgs boson using up to 4.9 fb \(^{-1}\) of \(pp\) collision data at \(\sqrt{s}=7\) TeV with the ATLAS detector at the LHC. Phys. Lett. B 710 (1), 49–66 (2012). https://doi.org/10.1016/j.physletb.2012.02.044

ATLAS and CMS, Collaborations: Candidate Higgs boson events from collisions between protons in the LHC. General Photo (2013). https://cds.cern.ch/record/1630222

CERN: ATLAS Experiment at CERN. http://atlas.cern/ . Accessed: 27-04-2023. (No date available)

CMS Collaboration: A measurement of the Higgs boson mass in the diphoton decay channel. Phys. Lett. B 805 , 135425 (2020). https://doi.org/10.1016/j.physletb.2020.135425

ATLAS Collaboration: Combined measurement of the Higgs boson mass from the \({H}\rightarrow \gamma \gamma\) and \({H}\rightarrow {ZZ}^*\rightarrow 4l\) decay channels with the ATLAS detector using \(\sqrt{s}\) = 7, 8 and 13 TeV \(pp\) collision data. Phys. Rev. Lett. 131 (25) (2023). https://doi.org/10.1103/physrevlett.131.251802

ATLAS Collaboration: Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC. Phys. Lett. B 716 (1), 1–29 (2012). https://doi.org/10.1016/j.physletb.2012.08.020

P.W. Higgs, Broken symmetries and the masses of gauge bosons. Phys. Rev. Lett. 13 , 508–509 (1964). https://doi.org/10.1103/PhysRevLett.13.508

Article   ADS   MathSciNet   Google Scholar  

CMS Collaboration: Observation of the Higgs boson decay to a pair of \(\tau\) leptons with the CMS detector. Phys. Lett. B 779 , 283–316 (2018). https://doi.org/10.1016/j.physletb.2018.02.004

CMS Collaboration, Measurement of the Higgs boson width and evidence of its off-shell contributions to \({ZZ}\) production. Nat. Phys. 18 (11), 1329–1334 (2022). https://doi.org/10.1038/s41567-022-01682-0

CERN: CERN Yellow Reports: Monographs, Vol 2 (2017): Handbook of LHC Higgs cross sections: 4. Deciphering the nature of the Higgs sector. CERN (2017). https://doi.org/10.23731/CYRM-2017-002 . https://e-publishing.cern.ch/index.php/CYRM/issue/view/32

LHC cross-sections working group: SM Higgs Branching Ratios and Total Decay Widths (update in CERN Report4 2016). https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageBR . Accessed: 16-04-2024 (2023)

G. Marchiori, Search for rare and lepton flavor violating decays of the Higgs boson with the ATLAS detector. https://indico.cern.ch/event/577856/contributions/3420203/attachments/1878247/3095626/20190711_-_Higgs_boson_rare_and_lepton-flavour-violating_decays_-_no_backup.pdf . Talk presented at the 2019 European Physical Society Conference on High Energy Physics, Ghent, Belgium, 10–17 July 2019 (2019)

CMS collaboration: Evidence for Higgs boson decay to a pair of muons. J. High Energy Phys. 2021 (1), 148 (2021). https://doi.org/10.1007/JHEP01(2021)148

ATLAS Collaboration: Search for the Dimuon Decay of the Higgs Boson in \(pp\) collisions at \(\sqrt{s}=13\) TeV with the ATLAS detector (2017). https://doi.org/10.1103/physrevlett.119.051802

ATLAS Collaboration, A search for the dimuon decay of the Standard Model Higgs boson with the ATLAS detector. Phys. Lett. B 812 , 135980 (2021). https://doi.org/10.1016/j.physletb.2020.135980

J.A. Brownlee, Gentle Introduction to XGBoost for Applied Machine Learning” Machine Learning Mastery. https://machinelearningmastery.com/gentle-introduction-xgboost-applied-machine-learning/ . Accessed: 15-04-2024 (2021)

C. Adam-Bourdarios, G. Cowan, C. Germain, I. Guyon, B. Kégl, D. Rousseau, The Higgs boson machine learning challenge, in Proceedings of the NIPS 2014 Workshop on High-energy Physics and Machine Learning. Proceedings of Machine Learning Research , vol. 42, ed. by G. Cowan, C. Germain, I. Guyon, B. Kégl, D. Rousseau (PMLR, Montreal, 2015), pp.19–55

CERN: What’s so special about the Higgs boson? https://home.cern/science/physics/higgs-boson/what . Accessed: 15-04-2024 (2018)

ATLAS Collaboration: Dataset from the ATLAS Higgs Boson Machine Learning Challenge 2014. CERN Open Data Portal http://opendata.cern.ch/record/328# (2014)

E. Wulf, Search for Randall–Sundrum Gravitons at the LHC: Recent Results from ATLAS. https://indico.cern.ch/event/129980/contributions/1350939/attachments/90398/129356/evan_wulf_RS_diphoton_dpf.pdf . Talk presented at the 2011 Meeting of the Division of Particles and Fields of American Physical Society , Brown University, 8–13 August 2011

CERN: Extra dimensions, gravitons, and tiny black holes. https://home.cern/science/physics/extra-dimensions-gravitons-and-tiny-black-holes . Accessed: 16-04-2024 (2018)

CERN: Supersymmetry. https://home.cern/science/physics/supersymmetry . Accessed: 16-04-2024 (2018)

C.G. Lester, D.J. Summers, Measuring masses of semi-invisibly decaying particle pairs produced at hadron colliders. Phys. Lett. B 463 (1), 99–103 (1999) https://doi.org/10.1016/s0370-2693(99)00945-4

Download references

Acknowledgements

We acknowledge the work of the ATLAS Collaboration to record or simulate, reconstruct, and distribute the Open Data used in this paper, and to develop and support the software with which it was analysed. We also acknowledge the technical support provided by the Rutherford Appleton Laboratory, and the support from the masters and graduate students of the Oxford Standard Model and Beyond group in developing the repository of training materials. Finally, we acknowledge the Institute for Research in Schools for their past and continuing support for this project, for connecting the materials with the students and teachers without whom the results presented in this paper would not have been possible.

The funding was provided by UKRI Public Engagement (Grant No. BB/T018534/1), University of Oxford, Merton College, University of Oxford, and Science and Technology Facilities Council (Grant Nos. ST/R002444/1, ST/S000933/1, ST/W000628/1, ST/X00600X/1). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission.

Author information

Alan Barr, Ynyr Harris, Julie Kirk, Emmanuel Olaiya and Richard Phillips have contributed equally to this work.

Authors and Affiliations

Department of Physics, University of Oxford, Parks Road, Oxford, OX1 3PJ, UK

Eimear Conroy, Alan Barr & Ynyr Harris

Particle Physics Department, Rutherford Appleton Laboratory, Harwell, Didcot, OX11 0QX, UK

Julie Kirk & Emmanuel Olaiya

Institute for Research in Schools, Wellcome Wolfson Building, 165 Queen’s Gate, London, SW7 5HD, UK

Richard Phillips

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Eimear Conroy .

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Conroy, E., Barr, A., Harris, Y. et al. Real particle physics analysis by UK secondary school students using the ATLAS Open Data: an illustration through a collection of original student research. Eur. Phys. J. Plus 139 , 781 (2024). https://doi.org/10.1140/epjp/s13360-024-05518-z

Download citation

Received : 20 May 2024

Accepted : 29 July 2024

Published : 02 September 2024

DOI : https://doi.org/10.1140/epjp/s13360-024-05518-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Advertisement

• Find a journal
• Publish with us
• Track your research