CQS

What 528 Hz Actually Does (And What It Doesn't)

2026-08-21

Someone left a comment under one of my videos last spring saying the piece had repaired their DNA. They meant it kindly. I thanked them and moved on, but the comment stayed with me, because I had spent that same week in my studio in Istanbul retuning a Juno patch up by a hair and listening to the same four bars maybe two hundred times, and at no point had DNA entered my thinking.

I've been making 528 Hz music for a while now. I get asked about it enough that a proper answer seems overdue. This is that answer, and it's going to disappoint anyone who arrived hoping for confirmation.

The claim, as you'll usually meet it

Search for 528 Hz and you'll find it called the love frequency, the miracle tone, the frequency of transformation. The specific claim, the one that travels furthest, is that 528 Hz repairs damaged DNA. Usually there's a mention of an experiment where DNA in a test tube was exposed to the tone and healed. Sometimes there's a line about genetic engineers using it in laboratories. Often the frequency is described as ancient, part of a lost medieval scale that the Church suppressed.

Four claims, then: it's ancient, it was suppressed, it repairs DNA, and there's an experiment proving it.

I went looking for the sources on all four. What I found was more interesting than a simple debunk, and less flattering than the story.

The number is from 1999

The six "Solfeggio frequencies" (396, 417, 528, 639, 741, 852) first appear in print together in a 1999 book called Healing Codes for the Biological Apocalypse, by Leonard Horowitz and Joseph Puleo. Puleo said he recovered them by applying a numerological method to verse numbers in the Book of Numbers, chapter 7. The method is digital root reduction: add the digits of a number until you're left with one.

Do it yourself. 396 gives 3+9+6=18, 1+8=9. 528 gives 5+2+8=15, 1+5=6. Every number in the set reduces to 3, 6 or 9, which proponents treat as proof of a hidden order, usually with a Tesla quotation attached.

Here's the thing about that. Exactly one third of all whole numbers reduce to 3, 6 or 9. It isn't a discovery, it's arithmetic. And the numbers in the set climb in steps of 111, which is why the pattern repeats. You've built the result into the construction.

The medieval part doesn't survive contact either. Guido d'Arezzo, an eleventh-century monk, took the syllables ut-re-mi-fa-sol-la from the first notes of a hymn to John the Baptist. That's the actual origin of solfege, and it's real history. But those syllables were never attached to fixed pitches; they were relative labels, movable, the same idea as do-re-mi in a singing class today. There's a harder problem too. The hertz didn't exist as a unit until the International Electrotechnical Commission adopted it in 1935. Reliable frequency measurement of any kind postdates the seventeenth century. There is no historical evidence that medieval solfege encoded any fixed pitch in hertz, and no way there could be: you cannot put a number on a frequency before anyone can measure one.

One more thing I couldn't get past. Horowitz calls 528 Hz "MI," from mi-ra-gestorum in the hymn. But mi is E. If 528 Hz were E, you'd need an A tuned to 352 Hz, which is nowhere near any tuning standard anyone has used. Elsewhere in the same literature 528 is called C. It can't be both.

The DNA experiment

This one deserves care, because there is a real scientist at the bottom of it and I don't want to be unfair to him.

By his own account, Glen Rein holds a 1983 PhD in biochemistry from the University of London and worked at Mount Sinai School of Medicine and later at Stanford University School of Medicine. These credentials appear in biographies he supplies and I could not confirm them independently. In the mid-nineties he did experiments where DNA in solution was exposed to various inputs, and measured the results with ultraviolet absorbance at 260 nanometres.

That measurement is the whole story. Absorbance at 260 nm tells you about the winding state of DNA: whether the double helix is more or less tightly coiled, which changes how much UV it soaks up. It's a standard, useful lab technique. It is not a measure of damage, and it is not a measure of repair. A change in absorbance means the molecule's conformation shifted. Reading that as "the DNA healed" is like reading a thermometer and announcing the patient's mood.

The work also wasn't published in peer-reviewed journals. It appears in conference proceedings from 1993 to 1996. And the frequently cited date of 1988 for a Rein music-and-DNA study? I looked, repeatedly, and could not find any such paper. As far as I can tell it doesn't exist.

Then there's the detail that made me sit back. Jonathan Goldman, one of the main popularisers of this material, describes Rein's music results on his own website and notes they came to him as a private communication, unpublished. In that account, a single 528 Hz tuning fork produced no measurable effect at all. Something was only observed with an ascending sequence of nine forks played together.

So the strongest proponent's own primary source says the thing that 528 Hz is famous for didn't happen.

One more thing worth knowing, since resonance is usually how people picture this working. DNA does have measurable vibrational modes: a 2016 paper in Nature Communications by Gonzalez-Jimenez and colleagues put them at 1.38 and 2.83 terahertz, billions of times higher than 528 Hz. So whatever might be happening, it isn't the molecule resonating with the note.

I want to be careful here, because "sound can't affect cells" would be too strong a claim to make. Cells do respond to mechanical stimulation, and that's an active research field. What's missing for 528 Hz specifically is a mechanism anyone has proposed, a published measurement of DNA, and robust replicated evidence of an effect unique to this frequency rather than to music in general.

The studies that do exist

There are peer-reviewed papers on 528 Hz. I read them so you don't have to, and I'd rather describe them accurately than either dismiss or oversell.

The most-cited is Akimoto and colleagues, 2018, testing 528 Hz against 440 Hz music for five minutes. Salivary cortisol fell, oxytocin rose, tension scores dropped. Real measurements, real result. The sample was nine people. Nine. With multiple outcomes measured at multiple timepoints and no correction for that, which is the statistical equivalent of taking many shots and reporting the ones that landed. No study has repeated its design, 528 Hz against standard 440 Hz tuning with salivary cortisol and oxytocin as the outcome, so the result stands unreplicated. The journal is published by Scientific Research Publishing, a company widely described as predatory, and the paper has no PubMed record.

There's a 2017 cell-culture study on astrocytes exposed to ethanol, reporting less cell death with 528 Hz. Its problem isn't size, it's design: no other frequency was tested. The comparison was sound versus silence. That tells you nothing about 528 specifically, only that the cells behaved differently with a loud speaker running.

The best study in this area is recent and much larger. Bozok and colleagues, 2026, in Brain and Behavior: 162 participants, randomised, three arms (control, 528 Hz, 432 Hz). Two salivary markers, BDNF and CREB, rose in the 528 Hz group; a third rose in the 432 Hz group instead. Cognitive performance on a Stroop test did not differ at all. I find this study genuinely interesting and I'd like to see it replicated. It is one study, it was a single session, and the markers it moved are contested proxies rather than clinical outcomes.

None of these peer-reviewed studies measured DNA damage or repair. The one set of experiments that did was never published, and measured winding rather than repair.

There's a parallel literature on 432 Hz with the same shape: small pilots showing borderline heart-rate differences, and then a 2025 randomised crossover trial of monochord sessions in 43 cancer patients comparing 432 Hz to 443 Hz. On the psychological outcomes, anxiety, stress, fatigue and well-being, both tunings helped and neither beat the other. The authors did report larger cardiovascular changes under 432 Hz, in a trial of 43 people that nobody has yet repeated.

What the evidence actually supports

Now the part I think gets buried, because it's less exciting and much more useful.

Music helps people sleep. This isn't fringe. It's a Cochrane review, Jespersen and colleagues, updated 2022, pooling thirteen randomised trials with 1,007 participants. Across the ten trials that measured sleep quality on the Pittsburgh Sleep Quality Index, music produced a mean difference of 2.79 points, which the reviewers rated moderate-certainty evidence. In Cochrane terms that is a meaningful result. Nightly listening, twenty-five to sixty minutes.

The review specifies no genre. No tuning. No frequency.

And a 2018 study in PLOS ONE asked 651 people what they actually listen to for sleep. The answers named 545 different artists across fourteen genres. Bach, Mozart, Ed Sheeran. The authors concluded that the choices reflect personal preference rather than any shared musical structure, and that the mechanisms people describe are relaxation, distraction from rumination, and habit.

That's the real finding, and it points somewhere specific. What seems to matter is arrangement: slow tempo, predictability, no sudden dynamics, nothing that snaps you back to alertness at minute forty. Music you've chosen and return to, so it becomes a cue. None of that is a property of pitch.

So how big a difference is 528 Hz anyway

Worth knowing, this. 528 Hz is a C, but only if your A is tuned to about 444 Hz rather than the standard 440. Standard C5 sits at 523.25 Hz, so 528 is the higher of the two by 4.75 Hz, or about 15.6 cents. Slightly sharp, in other words, not flat.

Fifteen cents is a real difference and I don't want to pretend otherwise. Play the two notes at the same time and you'll hear them beat against each other, roughly five beats a second. In a careful side-by-side test, trained ears can pick it out.

But that isn't how anybody listens to this music. Play the two versions a day apart, or put on an hour of ambient tuned one way with nothing to compare it to, and most listeners could not reliably identify which is which. Unless you have absolute pitch, which almost nobody does, there's no reference point. The whole piece is shifted together, so every interval inside it stays exactly the same.

There's a detail here I find genuinely funny. 432 Hz music is tuned about 32 cents below standard, and 528 Hz music sits about 16 cents above it. The two most popular healing frequencies pull in opposite directions from concert pitch. They can't both be the natural tuning that the modern world abandoned.

Why I still tune to it

Which brings me to the awkward part, since I've just spent two thousand words dismantling the marketing for something I make.

I tune to 528 Hz for three reasons, none of them medical.

The first is that a fixed reference is useful. Everything I make sits in the same tuning, so pieces recorded months apart can be layered without retuning, and my analog synths and the singing bowls I record get calibrated once against a known point. Bowls in particular don't come from the factory in concert pitch. Choosing a reference and holding it makes a catalogue coherent.

The second reason is aesthetic, and I'll state it carefully because it's the easiest place to fool yourself. Working at a fixed reference changes small things I notice over a long session: how oscillators beat against each other, how a long reverb tail settles. I wouldn't bet money on hearing it in a blind test, and I'm not going to claim the pitch itself carries anything. But I'm the one in the room for eight hours making decisions, and the constraint shapes what I make.

The third is that people arrive looking for it. Someone searching for 528 Hz music at one in the morning is looking for something to sleep to, and they've adopted this number as their way of finding calm, slow, careful music. I'd rather be honest with them and still be there than pretend the search doesn't exist.

If it turned out tomorrow that a piece I recorded at 440 helped someone sleep just as well, I would not be surprised. I'd expect it. The sleeping is done by the arrangement: the slow movement, the absence of anything sudden, the hour of not being interrupted. That's what I actually work on.

The number is where some people find the door. What's behind it was never the number.

Composed Quiet Studio makes original ambient music for sleep, meditation and focus, composed and performed by hand in Istanbul. This article is about music and about evidence; it isn't medical advice, and music isn't a treatment for insomnia or any other condition.

Sources

More from the blog

Hear it in practice

Deep Meditation (528 Hz) album cover
Deep Meditation (528 Hz)Meditation · 528 Hz
Album ›

New release alerts

One short email when a new release is out. No noise, and you can leave with one click.

You will get a confirmation email first. Details in the privacy notice.