Industries / Consumer goods / Candles and home fragrance
Diffusers, sprays and incense
This work makes the scented home products that are not candles: aerosol and pump sprays, plug-in and electric diffusers, reed diffusers, wax melts and warmers, and incense sticks and cones.
7
Milestones achieved, from 1943 to 2025.
3
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1943
A wick in a bottle
Air Wick launched in the United States in 1943 as a chlorophyll-based liquid in a glass bottle, released into the room through a wick, by Reckitt's account. It set the pattern for air fresheners that work by steady evaporation, the idea behind today's reed diffuser.
LessonThe early claim that chlorophyll removed odours did not hold: a paper in the Journal of the Society of Cosmetic Chemists found that natural chlorophyll does not dissolve in water and showed no deodorant action in water except at high alkalinity.
Credit
Air Wick, later a Reckitt brand
1956
Air freshener goes aerosol
S.C. Johnson launched Glade in 1956 from the new products division led by Sam Johnson, and a company history says it "featured advanced technology in aerosol manufacturing". The spray can became a standard format for air care.
Credit
S.C. Johnson & Son, Racine, Wisconsin; Samuel C. Johnson
1990
Plug-ins run all day
Glade Plug Ins, scent dispensers plugged into a wall socket, came out in 1990 according to a company history in Gale's encyclopedia, which says they "quickly gained the top position in the continuous-action air freshener category"; a Mother Jones timeline gives 1989 for plug-ins in general. Heat now drove scent into the room around the clock, with no one pressing a button.
LessonAlways-on products made scent a constant indoor emission, which later chamber tests measured (see Fragrance, 2006).
Credit
S.C. Johnson & Son
2007
Phthalates found in fresheners
The Natural Resources Defense Council (NRDC) tested 14 common US air fresheners and found phthalates, chemicals that can disturb hormones, in 12. None listed phthalates on the label, including products sold as "all-natural" and "unscented". Walgreens announced the next day that it would pull its own-brand air fresheners, and in March 2009, NRDC reported, S.C. Johnson said it would list ingredients for its air care products and phase out its remaining phthalates by 2012.
LessonOutside testing changed formulas and labels faster than regulation did.
Credit
NRDC (Gina Solomon, Mae Wu); Walgreens; S.C. Johnson
2008
Incense smoke tied to cancer
The Singapore Chinese Health Study followed 61,320 adults and found that long-term incense users had 1.8 times the risk of squamous cell cancers of the respiratory tract (95% confidence interval 1.2 to 2.6). A US EPA team had earlier measured fine-particle emissions of 5 to 56 mg per gram burned across 23 types of incense (2002).
LessonA 2015 lab study at the South China University of Technology, co-run with a tobacco company, found smoke from two incense types more toxic to cells than the cigarette it tested, and its lead author warned against concluding that incense is generally worse than cigarettes.
Credit
Jeppe Friborg, Jian-Min Yuan, Mimi Yu and colleagues; James Jetter and colleagues, US EPA
2019
India restricts incense imports
India's Directorate General of Foreign Trade moved agarbatti, incense sticks, and other burned scent products from "Free" to "Restricted" imports by a notification of 31 August 2019, so importers need a licence. In June 2020 the Finance Ministry raised the duty on round bamboo sticks, the core of most agarbatti, from 10% to 25%. The government's Khadi and Village Industries Commission then funded local stick-making machines, saying India used about 1,490 tonnes of agarbatti a day but made only 760 tonnes (its figures).
LessonTrade rules, not technology, reshaped India's incense supply chain.
Credit
Ministry of Commerce (DGFT); Ministry of Finance; Khadi and Village Industries Commission
2025
Flame-free melts still pollute
Wax melts are scented wax melted in an electric warmer, with no flame. They spread through direct sales firms such as Scentsy, relaunched in Idaho in 2004 and counting more than 100,000 sellers by 2011, the Deseret News reported. Purdue University engineers measured them in a test house: wax melts gave off far more monoterpenes, a class of scent molecules, per gram of wax than candles did (about 100 mg per gram), and these reacted with indoor ozone to form bursts of more than a million nanoparticles per cubic centimetre, comparable to burning candles and gas stoves.
LessonNo flame does not mean no particles.
Credit
Nusrat Jung, Brandon Boor and colleagues, Purdue University (Liu and colleagues, 2024; Patra and colleagues, 2025)
Watch for
India sets an agarbatti standard
India released IS 19412:2025, a Bureau of Indian Standards specification for incense sticks. It bans the insecticides alethrin, permethrin, cypermethrin, deltamethrin and fipronil and the chemicals benzyl cyanide, ethyl acrylate and diphenylamine, sorts sticks into machine-made, hand-made and masala types, and lets products that meet it carry the BIS mark. The government puts the industry at about ₹8,000 crore a year.
Achieved whenBIS lists makers licensed to the standard, and independent tests of marked sticks bought in shops find none of the banned substances.
Claimed by
Ministry of Consumer Affairs and the Bureau of Indian Standards, December 2025
Two-scent waterless diffuser
SCENTA, a Chinese maker of scent devices, launched the A-T302 at CES 2026, a diffuser that turns scent oil into a fine mist without water through two nozzles and switches between two scents from a phone app. It calls itself "the No. 1 global seller of two-fluid scent diffusers in 2024".
Achieved whenindependent market data confirms the sales claim, or an independent test publishes output and switching times.
Claimed by
SCENTA, January 2026
App-run diffuser with vials
Shelly launched its Cury diffuser in Europe at about €83, with two scent vials, strength and schedules set from an app, and voice control. Shelly says each vial lasts about 120 hours of average use, and that one scent protects against mosquitoes.
Achieved whenan independent test confirms vial life at stated settings, and the mosquito claim passes the kind of efficacy test required of insect repellents.
Claimed by
Shelly, February 2026
Open
01
Reed diffuser evaporation dataset
An open dataset of how fast reed diffusers lose liquid, in grams per day, by reed type, number of reeds, base liquid and room temperature, used to test the "lasts up to X weeks" claims on shop-bought diffusers.
The briefCredit · unclaimed02
Test low-smoke incense claims
A published protocol and results per product for PM2.5, fine particles, given off per gram burned, comparing "low-smoke" incense with each maker's regular line, from Japanese and Indian makers.
The briefCredit · unclaimed
Open 01 · the brief
Reed diffuser evaporation dataset
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open dataset of how fast reed diffusers lose liquid, in grams per day, by reed type, number of reeds, base liquid and room temperature, used to test the "lasts up to X weeks" claims on shop-bought diffusers.
Why it is still open
The only measured rates found are in a maker's patent for a competing wick: a Stonewick patent, filed in 2008 and granted in 2013, reports natural reeds at about 0.09 g per hour per reed against 0.25 g per hour for its own wicks. Searches of Europe PMC (17 records, mostly poisoning reports), Zenodo, Hugging Face, Kaggle and GitHub found no open evaporation data.
Built with what exists
1 kg load cells with HX711 amplifiers on an ESP32, or lab balances with a serial output from makers such as Ohaus or A&D; Sensirion SHT45 temperature and humidity sensors; Python for logging and analysis; Zenodo for the data. An AI model that designs working circuit boards, such as Claude Fable 5.2, can lay out a logger board for many load cells (see the Note "Three working devices"). Diffuser liquids are kept away from children, pets and flames.
First step
Build a four-channel logger, check it with known weights, and start two identical shop-bought diffusers, one with 4 reeds and one with 8.
Done looks like
At least 24 diffusers logged every 10 minutes for 60 days, covering 3 reed types, 2 base liquids and 2 room temperatures (for example 20 °C and 26 °C), including at least 10 shop-bought products with stated lifetimes, with measured life divided by claimed life reported for each.
Open 02 · the brief
Test low-smoke incense claims
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A published protocol and results per product for PM2.5, fine particles, given off per gram burned, comparing "low-smoke" incense with each maker's regular line, from Japanese and Indian makers.
Why it is still open
Makers' smoke claims are in words or relative numbers. Nippon Kodo groups its line as "Less Smoke", "Lesser Smoke" and "Little to None"; a Shoyeido line is sold as giving "30% less smoke than our traditional incense". A National Taiwan University researcher told TVBS in June 2026 that eco-friendly and smokeless products his team tested for Taipei's Longshan Temple produced pollutants that were "virtually unchanged", and no product-level data from that work was found. The US EPA's measurements of 23 incense types (2002) came before the current low-smoke lines.
Built with what exists
Sensirion SPS30 particle sensors in a box of about 1 m³ with a fan venting outdoors at a measured air change rate, or a lab fume hood; filter samples on PTFE filters with a small sampling pump, weighed on a university microbalance, to correct the sensors; a 0.001 g balance to weigh each stick before and after; Python for the emission calculation. Ordinary consumer incense is burned attended, with normal fire precautions.
First step
Buy a box of Shoyeido's low-smoke incense and its traditional counterpart, and run three burns of each in a vented box with one SPS30, weighing the sticks before and after.
Done looks like
At least 10 matched pairs (low-smoke and regular from the same maker), 3 sticks each, with PM2.5 emissions in milligrams per gram; a verdict per pair on whether, and by how much, the low-smoke product cuts emissions, set against any stated figure such as 30%; protocol, data and code published.