Episode 52

Beyond Silicon: Perovskite Power for the Next Generation of LoRaWAN - Josh Douglas

This episode looks at how perovskite photovoltaics can reshape the power budget behind an IoT data platform for smart cities. Josh Douglas explains why conventional silicon solar cells struggle with indoor light, off-angle mounting, and rigid glass, and how thin perovskite nanomaterials unlock more flexible, energy-dense LoRaWAN node designs. We connect better energy harvesting to iot data integration platform tools, higher message frequency, and richer sensing that supports edge AI and “physical AI” workloads. You’ll also hear practical details on electrical compatibility, iot data ingestion best practices, and how perovskite modules can often drop in for small silicon panels. The conversation closes with commercialization realities, distribution through familiar channels, and what this means for long-lived devices feeding smart city data platforms. Listen to learn how improved power translates into more capable LoRaWAN deployments and smarter urban infrastructure.

Josh on LinkedIn

CPTI

  • Helium Global IoT Coverage - Want to know if Helium coverage exists where you need it? Check out this map!
  • Helium Foundation - The Helium Foundation's IoT Working Group (IOTWG) has generously provided support for the first 6 months of shows, please go check them out and consider using the Helium LoRaWAN as a primary or backup on your next deployment. With over a quarter million gateways deployed worldwide, it's likely that you have and can use Helium coverage.
  • Support The Show - If you'd like to support the MetSci Show financially, here's where you can donate on a one-time or an ongoing basis. Thank you!
  • MetSci Show - If you'd like to use our IoT or AI Data Value calculators, or you'd like to contact me, the MetSci Show site is the best way to do it.
  • MeteoScientific Console - Use LoRaWAN - The MeteoScientific Console allows you to use LoRaWAN today. As long as you have Helium coverage (and you probably do, about 90% of populated areas in the world have a gateway within 2 miles), you can onboard a sensor. You can always check coverage at https://explorer.helium.com and switch to the "IoT" tab in the top right.

Transcript
Speaker:

Today's guest

2

:

on MeteoScientific's

The Business of LoRaWAN is Josh Douglas,

3

:

COO and co-founder of CPTI, a company

building perovskite

4

:

based energy harvesting designed to change

how we power connected devices.

5

:

We met at CES, where Josh showed me

a nanomaterial solar layer, thin enough

6

:

to go on almost any surface and capable

of harvesting energy from indoor light,

7

:

off angle light, and wavelengths

traditional silicon struggles to capture.

8

:

In this conversation, we talk about what

perovskites actually are,

9

:

why conventional silicon panels

leave energy on the table,

10

:

and how higher energy density

can unlock more frequent reporting,

11

:

additional sensing,

and even edge AI in LoRaWAN deployments.

12

:

We also get into the real world issues

lifetime limitations, commercialization

13

:

hurdles, and how CPTI is bridging the gap

from lab science to digital listings

14

:

and deployable hardware.

15

:

This episode is sponsored

by the Helium Foundation

16

:

and is dedicated

to spreading knowledge about LoRaWAN.

17

:

If you'd like to try Helium’s

publicly available global LoRaWAN for free

18

:

and support

the show, sign up at metsci.show/console.

19

:

Now let's dig in the conversation

with Josh Douglas.

20

:

Josh,

thanks so much for coming on the show.

21

:

Welcome. Thank you. Good to be here.

22

:

I'm excited to have you on.

23

:

We're going to talk about perovskite.

24

:

We met in Las Vegas at CES.

25

:

And you had

26

:

show me this pretty cool material

that I think of as like a solar panel.

27

:

You can maybe put on anything,

but I thought we'd start with the

28

:

tell me a little bit more about what

this stuff is, and why people in IoT

29

:

and LoRaWAN might want to pay attention

to it.

30

:

Totally. What this stuff is. So.

31

:

So what

what we're building at CTI is perovskite

32

:

based energy harvesting perovskite.

33

:

It's a family of materials.

34

:

It's really

35

:

a few different kinds of materials,

but it's most infamous for solar cell

36

:

spectrum for for photovoltaics

and current photovoltaics.

37

:

Current solar,

which I know you're quite used to.

38

:

Nick is really good at, I think, cheap.

39

:

Frankly, at current solar, it's

very cost efficient

40

:

and it's

kind of built this industry around it,

41

:

but it is is fairly inefficient

at actually harnessing energy.

42

:

It only is responsive

to basically one wavelength of light

43

:

and needs to absorb that light,

at a pretty direct angle.

44

:

It needs to be pretty spot on to capture

anything material, which I see you nodding

45

:

your head.

46

:

So this is a problem

that I know that you see.

47

:

Yeah.

48

:

Furthermore, current silicon pretty much

needs to be adhered on to glass.

49

:

That's how you really, really harness it.

50

:

So what that means is, as you deal

with, with a cheap certainly,

51

:

but a pretty inefficient material

that has pretty extreme limitations.

52

:

What we're working with, with perovskite,

really, it really changes that dynamic.

53

:

And so as we talk about and I was I can

54

:

I can speak more to what's so cool

about perovskite a few things.

55

:

One is it can really harness

any kind of light.

56

:

So whether it's outdoor light

like in the sun,

57

:

whether it's cloudy

light or even indoor light,

58

:

we can turn all of that

to energy at any incident angle.

59

:

So even if the sun's rising or setting,

even if it's off angle light,

60

:

if it's been bouncing off

a few different substances,

61

:

we can still capture that and turn

that into energy, which is amazing.

62

:

And on top of that, perovskite, as I know

we discussed it's a nanomaterial.

63

:

So it's thinner than a human hair.

64

:

It's very energy dense.

65

:

We can put it really get on any surface

with with minimal added mass.

66

:

Okay.

67

:

So let me make sure I'm understanding that

if I think of solar panels

68

:

and I put up a bunch on these kind of off

grid LoRaWAN gateways

69

:

is it's a big, big piece of glass,

you got to be careful with it.

70

:

Yeah. Relatively inefficient.

71

:

I mean, we're more efficient than when

we started with solar panels in the 50s.

72

:

But still, I forget what it is at now.

73

:

Maybe 20% ish, maybe less. Yeah.

74

:

And then a little bit heavy.

75

:

And then it has to have the sun

hitting it directly.

76

:

So I've always had to basically match the,

the angle to the latitude

77

:

is what I think it

or the way I think of doing it.

78

:

And so with this stuff with perovskite,

you don't have to do any of those

79

:

and you can put it on anything.

80

:

Is there?

I mean there's got to be a downside.

81

:

Is this thing super expensive.

82

:

Like what's the

why would I not use perovskite.

83

:

Why you would not use it is

because it just really hasn't been ready.

84

:

I wouldn't

call it a new class of material,

85

:

but it's one

that's been in the pipeline for a while.

86

:

It's been in the works for some time.

87

:

I did it work on for off site in undergrad

and really it's it's

88

:

come a long way since then

and namely in its performance,

89

:

its efficiency, its lifetime,

all these different things.

90

:

It's it's improved.

91

:

And so the reason why

you haven't used perovskite.

92

:

Yeah.

Just because it hasn't been ready yet.

93

:

We're now beginning to see companies

really commercialize perovskite.

94

:

CPI really leading the way.

95

:

You know what what I will say

to your point is that cost, because it's

96

:

a newer material goes on that

same industry around it like the silicon,

97

:

but because it's an inert material,

it's pretty easy to process that.

98

:

Over time, you will see the the cost fall

dramatically.

99

:

Okay. And what's the cost comparison?

100

:

Now, if I'm buying a whatever,

a little solar panel that I might use

101

:

for an IoT device. Yeah,

a couple inches by a couple inches. Yeah.

102

:

What does it

103

:

same thing look like as far as the energy

that I need to match it in perovskite.

104

:

Yeah, totally.

105

:

So so right now if you say buy or sell off

106

:

digitally, for example, or aero,

our modules are on a surplus

107

:

energy as an be, you know, like like what

we're capturing above and beyond.

108

:

We are getting you anywhere from 5 to 20%

109

:

more energy, more efficiency.

110

:

What that means in practice,

because as you said, most cells currently

111

:

are on the order of 10 to 20% efficient.

112

:

So, so so we are again

depending on your form factor on

113

:

on your latitude or your conditions,

we are upwelling the energy that you're

114

:

that you're able to harness

so we can drive pretty awesome gains

115

:

in energy, which allows you to record

more frequently, which allows you to pack

116

:

more sensing on your node,

which really opens up

117

:

all these different application spaces,

which we haven't seen before.

118

:

I mean, the obvious one is AI

is this giant energy sucking machine.

119

:

What's that? I've got I've got it. Yeah.

Oh yeah.

120

:

That's like, oh, wow, man,

this guy's really been in the closet now.

121

:

So to this AI piece,

it seems like this would fit

122

:

well into the narrative by saying, hey,

you can either use a smaller panel

123

:

because this stuff is more efficient,

or use the same size panel

124

:

and kind of replace what you've got,

and now you've got more energy to use.

125

:

Is that the general storyline here,

126

:

or is there additional pieces to it

that I might be missing just because I'm

127

:

coming into this

for the first time? Totally.

128

:

I mean, you really hit the nail

on the head, which is that

129

:

energy is critical.

130

:

And in many cases

we are stuck with batteries.

131

:

We're stuck with outdated solar.

132

:

And so we've we've really grown

used to these energy constraints.

133

:

Right?

134

:

I like to think of

to go back a year or two or a few decades

135

:

to the Walkman versus the iPod. Right.

136

:

And how the unlock with the iPod

was that you

137

:

carry all these songs in your pocket

that you couldn't before,

138

:

and most people at first

didn't see a need for that.

139

:

Right.

140

:

They thought that carrying

all these albums around in your pocket,

141

:

what's the point

now where you take it for granted? Right?

142

:

That's how we feel with energy

and that we've we don't know how much

143

:

we've been limited because we think in a,

in a very different framework.

144

:

We want to change that paradigm. Right.

145

:

And we know when it comes

to certainly latest, greatest

146

:

machine learning AI,

that's a big driver, right.

147

:

When we talk about edge compute

and we talk about physical AI,

148

:

these are very real, real tailwinds

that we're seeing across Lora,

149

:

that we're seeing across devices,

something that would be a big tailwind.

150

:

But just even in general, right?

151

:

I mean, even if you aren't

adding compute on the edge,

152

:

you want to collect more data,

you want to create.

153

:

And I mean, I'm

I'm based in Silicon Valley,

154

:

so I'm in this world like, you know,

we want to create these role models.

155

:

We want to add autonomy.

156

:

You want to create these frameworks

such that you can add intelligence

157

:

all over the place,

which will take a lot of sensing, a lot of

158

:

data on the C, a lot of computer places,

and all these things take more energy.

159

:

The irony, of course, with all this,

Nick, is, we are surrounded by energy,

160

:

the form of light.

161

:

There's tons of energy all around us,

162

:

but we don't harness it

due to the limitations of our materials.

163

:

Yet this is this is how we figure out CPI,

164

:

and this is what we're really looking

for it to bring to the lower community.

165

:

Okay, cool.

166

:

And if I think about it,

you can't be just a material science nerd.

167

:

You've got to put the whole thing together

and be able to say, hey,

168

:

this is what this material is,

this is how it works.

169

:

And then here's the business case

170

:

and here's how you buy it,

and here's how you install it.

171

:

There's a bunch of different things

you have to do.

172

:

You've worked at Apple in the past.

173

:

So hopefully some of that

174

:

like hey, this thing has to work

easily got imbued into you.

175

:

Talk me through the bridging

the gap between this kind of cool material

176

:

that works and then getting it deployed

in the real world at scale.

177

:

Yeah. That's critical.

178

:

That is what holds back

many exciting innovations across

179

:

science is how to actually bridge

that gap.

180

:

Right?

181

:

Even for for me,

throughout my my own career.

182

:

Again,

I did work on perovskites undergrad.

183

:

I went in thinking I would go off

and get a PhD in material science

184

:

to save the world.

185

:

Learning about perovskites

moved me away from that, not because I

186

:

lacked faith in the material,

but just because I saw firsthand.

187

:

While there was great work happening

in the lab, the actual translation,

188

:

the actual movement from lab to

189

:

market was still unsolved

and still very murky.

190

:

And that's again,

that's how I think about it at CTI.

191

:

And we we have the the opportunity to work

192

:

with these really game

changing materials in perovskites,

193

:

having these best in class chemistries,

really working on the cutting edge.

194

:

And there are so many places to

to put it right.

195

:

And I mean,

we could be on a different podcast

196

:

talking about the opportunity

for perovskites in space on satellites,

197

:

the opportunity

for perovskites on the walls of buildings,

198

:

or of course, in solar farms,

the classic polar application.

199

:

All these markets

have their their own merits and demerits.

200

:

And I'd be happy to talk about those

if you're curious.

201

:

But what what makes us so excited

about again,

202

:

like sensing in LoRaWAN

is that it's energy where you need it.

203

:

And that's where we can make a difference.

204

:

And so for us, it's a matter

of both understanding the material

205

:

but also understanding of the spaces

in which we could apply it, understanding

206

:

how customers could adopt it,

how they can integrate it in,

207

:

and really making it as easy as possible

for people to adopt.

208

:

We don't want to add all these barriers

and gateways.

209

:

We think a lot about how to distribute,

how to get this into people's hands,

210

:

or how to talk about this

in the first place, such that we can get

211

:

people excited and educated

and not be some research project.

212

:

I think about the the guests who have

come on Julian Berlin over at Volvo or,

213

:

you know, from

214

:

from that kind of really high level, like,

hey, these guys are rolling this out

215

:

across giant campuses

to Robert Boggs in Germany, who's

216

:

rolling out in his village to Dean Marsh

in the UK with those guys are listening.

217

:

Where would they go to get perovskite

for their next project and see, like, hey,

218

:

is this going to work?

219

:

Can you just ordered on digit key

220

:

and it's like a swap out with your current

solar panel.

221

:

Is it some different weirdo voltage?

222

:

Is it some like what are the things

that they're going to get the gotcha from?

223

:

And then one of the things where I'm like,

what did I know about this before?

224

:

Yeah, yeah, definitely.

225

:

It's a bit as I was saying prior,

Nick is just that this is a new material

226

:

and that it's really taken time

for industry to meet up with the science

227

:

and to find a way to, to commercialize it

at a we're leading the pack.

228

:

We have undergone certifications.

229

:

We've undergone the testing needed such

that folks can use this.

230

:

We're being listed on the key on arrow.

231

:

All the places that you would want

to purchase are these things, you know,

232

:

and and that's something that

that we think about differently

233

:

than other people is just really trying to

to short that leap.

234

:

Of course, folks are always

235

:

welcome to reach out to us

via our website in the interim.

236

:

And this is just Icom.

237

:

Sibeko. Exactly.

238

:

Dukkha. Okay. Yeah.

239

:

And then when it comes to the voltages,

240

:

the voltages, everything, it's very,

very similar to, to silicon.

241

:

That's so different.

242

:

The thing that, again, really sets

perovskite apart is being able to capture

243

:

different wavelengths of light and

different energies at different angles.

244

:

And so generally with silicon, right,

like on the of curve of the energy

245

:

that it's capturing and what you're using,

generally you really only need

246

:

to capture the peak loads of energy

because it's a it's

247

:

a very steep drop off with silicon

perovskite,

248

:

we actually are more efficient

and lower light conditions.

249

:

And so we are able to harness more energy

and basically makes

250

:

makes more sense to harness energy

at a, at a wider range of voltages.

251

:

And currents and wattage and so on.

252

:

So it's really fast.

253

:

The only, the only difference is not like

this is a whole different paradigm.

254

:

You know, it, it can still work

with the same kind of control.

255

:

Or is the same kind of mix.

256

:

It's just that due to the material

you are inclined to

257

:

to harness more of the energy

and use it for your devices.

258

:

The interesting.

259

:

What's the simplest possible case

where someone would say like, hey,

260

:

I'm just going to swap out this stuff.

261

:

I've got 100 or 1000 or ten or whatever

it is.

262

:

Yeah. What would you say? Like,

hey, just start with this.

263

:

That way you get your feet wet

without getting in over your.

264

:

Yeah, at your neck. Totally.

265

:

That's it. Right.

266

:

Is trying it out.

267

:

And in many cases

you don't need a controller.

268

:

In many cases you can just swap this in

for for your silicon panels.

269

:

If you have some sort of PMI, C or mPPT

that's converting

270

:

voltage that that will will harness

much of the energy as well.

271

:

And then over time there's there's

obviously ways to optimize.

272

:

We're actually working

273

:

on our own reference designs on our own

like dev kids to again

274

:

make it easier for people to adopt

and do harness the RG.

275

:

But there are tons of ways

to optimize and perfect

276

:

and to harness every micro what you know,

but not every application needs that far.

277

:

We don't all need micro watts,

although some of us I say

278

:

that this is the world that does. Yeah,

yeah yeah yeah, for sure.

279

:

I know that with my little balloon project

280

:

we're trying to harvest every piece we can

because it's such a tiny little board.

281

:

Yeah. Let's see, is there

are there any big downsides?

282

:

Like how long does it last?

283

:

Does this thing just crumble

in five years?

284

:

And it's like,

oh, it's cheaper and better,

285

:

but you got to replace every five years.

286

:

Are there any other kind of big downsides

that folks should know about before

287

:

they go racing off to buy out

the entire stock of Digital Hero?

288

:

Yeah. You guys have. Yeah. Of course.

289

:

So so that has been the limiting factor

for perovskite is lifetime.

290

:

So what I was really yeah.

291

:

So back in the day

when when I was a researcher,

292

:

it was on the order of about a month

or two was the lifetime of perovskite.

293

:

Wow. Yeah.

294

:

It was very, very quick.

295

:

It would degrade very quickly.

296

:

Now it's up to around ten year

lifetime okay.

297

:

And then we are best in class at around

15 years lifetime.

298

:

So okay.

299

:

So this is more than enough

for most applications.

300

:

And obviously to

we are often sitting alongside a battery.

301

:

Right. We are harvesting energy

to work with a battery.

302

:

Most batteries are lasting for five

maybe ten years.

303

:

So you don't need to last for 20 years

if the battery is going to fail in ten.

304

:

And anyway, yeah, yeah, that makes sense.

305

:

It's interesting.

306

:

I didn't know

is so short at the start of it.

307

:

Although, yeah, I would say

308

:

if you're listening to this, don't

remember the one month thing.

309

:

Remember the 15 year thing.

310

:

That's the key here.

311

:

I always thought it was cool when people

ask about how long the solar panels last.

312

:

We say

313

:

we don't know because the first ones

314

:

that went up in the late

50s are still running.

315

:

Yeah, but, as you said,

if you're trying to charge a battery

316

:

that died 40 years ago, then who cares?

317

:

Yeah. Okay.

318

:

Well, well, that's like

also the dirty secret of silicon

319

:

of basically every other celebrity. True.

320

:

Is that they also have a lifetime

in a similar range,

321

:

maybe the lasting for 20 years

with only 10% degradation,

322

:

but they are covered under warranty

for 30 years

323

:

for whatever is needed to bank a project

to finance a project.

324

:

And what what happens is that these solar

producers, they sign these warranties,

325

:

thinking

326

:

and planning for the price to drop so much

327

:

that once the customer comes

328

:

knocking and collect on the warranty,

that it'll cost them

329

:

very little to to substitute

out the panels for the blue ones.

330

:

So even though they say 30 years,

that's just a marketing gimmick.

331

:

It's a very funny industry, Lives learn.

332

:

It's it's a nice little gamble

for a business to play.

333

:

Like, hey, by the time we replace these,

it's going to be cheaper.

334

:

Don't worry. Exactly, exactly. Ripping.

335

:

Well, Josh, thanks so much for carving out

time. I know you're busy.

336

:

I know the whole founding thing

is not like a 9 to 5.

337

:

It's more like 12 to 12.

338

:

So thanks for coming on

and talking to us about, Prescott.

339

:

Of course. Thank you for having us.

340

:

That's it for

this episode of The Business of LoRaWAN.

341

:

If you want to go deeper

and actually deploy devices,

342

:

the MeteoScientific

console is the fastest way to do that.

343

:

And honestly, it's

also the best way to support the show.

344

:

When you use the console, you're not just

listening, you're participating

345

:

in the same real world LoRaWAN work

we talk about here every week.

346

:

You can get started with the free trial

at meteoscientific.com.

347

:

Huge thanks to the sponsor of this show,

the Helium Foundation,

348

:

for supporting open LoRaWAN

infrastructure worldwide.

349

:

Check them out at helium.foundation

and if the show has been

350

:

useful, a quick rating or review

on Apple Podcasts or wherever you listen.

351

:

This really helps

352

:

people find it and helps the show grow

so we can help more people.

353

:

I'm Nik Hawks with MeteoScientific.

354

:

I'll catch you on the next episode.

All Episodes Previous Episode

About the Podcast

Show artwork for The Business of LoRaWAN
The Business of LoRaWAN
Learn From the Pros

Listen for free

About your host

Profile picture for Nik Hawks

Nik Hawks

Incurably curious, to stormy nights and the wine-dark sea!