Glenn Williams, Jeff Brum, Rachel Forcino, Fred Vogt

GlaxoSmithKline

XRPD Line Broadening Analyses to Study Micronization Induced Disorder and Environmental Annealing

Why Study Micronization Induced Disorder?

Non-Micronized Milling

Micronized

Process of micronization results in significant attrition with consistent micron size particles, but the high energy process may also induce an

element of “disorder” in the material.

While this process improves dissolution and particle uniformity of the API the resulting “disorder” may affect stability and physicochemical

properties, ultimately with a potential effect on dissolution and particle uniformity??

Micronization Mill

Changes in Physical Properties

Are we investigating all the possible changes that may occur and the impact of those changes on the final formulation/processing and product?

-0.14

-0.12

-0.10

-0.08

Heat F

low

(W

/g)

80 100 120 140 160 180 200

Temperature (°C)

Sample: GW786034B (Pazopanib) Bx 108001Size: 4.1240 mgMethod: STANDARD DSC METHODComment: Top of keg. Keg 2 of 8

DSCFile: I:...\GW786034B\108001 TOP KEG DSC.001Operator: COFRun Date: 16-Sep-2010 16:58Instrument: DSC Q2000 V24.2 Build 107

Exo Up Universal V4.2E TA Instruments

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

Heat F

low

(W

/g)

80 130 180

Temperature (°C)

Sample: GW786034B_K074440_DSCSize: 2.5840 mgMethod: DSC_GW786034Comment: EE176981

DSCFile: I:...\GW786034\GW786034B_K074440_DSC.001Operator: RGFRun Date: 07-Jun-2007 14:06Instrument: DSC Q1000 V9.0 Build 275

Exo Up Universal V4.2E TA Instruments

0.0

0.2

0.4

0.6

De

riv.

We

igh

t (%

/°C

)

70

75

80

85

90

95

100

We

igh

t (%

)

0 50 100 150 200 250 300

Temperature (°C)

Sample: GW786034B Ball MilledSize: 10.3660 mgMethod: Standard MethodComment: Ball Milled

TGAFile: GW786034B_US061115667_BALLMILLED_TGA.001Operator: BLRun Date: 01-Jun-2006 10:48Instrument: TGA Q500 V6.3 Build 189

Universal V4.2E TA Instruments

0.0

0.2

0.4

0.6

De

riv.

We

igh

t (%

/°C

)

75

80

85

90

95

100

105

We

igh

t (%

)

0 50 100 150 200 250 300

Temperature (°C)

Sample: GW786034B_K074440Size: 27.5940 mgMethod: Standard MethodComment: EE176981

TGAFile: I:...\GW786034\GW786034B_K074440_TGA.001Operator: MBRun Date: 07-Jun-2007 15:47Instrument: TGA Q500 V6.3 Build 189

Universal V4.2E TA Instruments

Surface Area

Surface Energy

CrystallinityXRPD

ThermalDSC, TGA

Moisture Sorp.GVS

“Disorder”?

Disordered

FRAC

Non-

crys

tallin

e

TURED

Crystal “healing” or environmental “annealing” is often observed in aged/environmentally exposed materials

Aging - API “hold times”

Determination of the true contributions/components of disorder is necessary to understand and predict the effects on physicochemical

properties and pharmaceutical processing

180.01°C

175.36°C(I)

158.83°C

-1.05

-1.00

-0.95

-0.90

-0.85

Hea

t F

low

(W

/g)

-6

-4

-2

0

2

Hea

t F

low

(W

/g)

-100 0 100 200 300 400

Temperature (°C)Exo Up Universal V4.2E TA Instruments

Micronized

Ball milled

Ball milled

Thermal Analysis Comparison of Milled API

Tg

Tg of amorphous standard higher than exotherm in micronized API

Surface vs Bulk Amorphous

Wu, T.; Sun, Y.; Li, N.; de Villiers, M.; Yu, L. Langmuir 2007, 23, 5148

Lei Zhu & Janan Jona & Karthik Nagapudi & Tian Wu Pharm Res (2010) 27:1558–1567

Sayantan Chattoraj, Chandan Bhugra, Chitra Telang, Li Zhong and Zeren Wang, Changquan Calvin Sun. Pharm Res, 2012, Volume 29, Number 4, Pages 1020-1032

Amorphous griseofulvin 13C T1 and T1ρ measurements

Lower 13C T1ρ and T1 values are consistent with higher molecular mobility in the ground sample

Variable temperature measurements were not performed because of “ground” sample instability – recrystallization occurs rapidly under MAS spinning conditions

13C T113C T1ρ

Melt-quench (initial) 1.60 ± 0.12 s 78 ± 13 ms

Melt-quench(ground) 0.74 ± 0.18 s 45 ± 13 ms

Griseofulvin 13C CP-TOSS spectra

13C CP-TOSS spectra

Micronized API(red trace) shows the signature of amorphous material

145.68°C

153.49°C

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

Hea

t F

low

(m

W)

-0.8

-0.6

-0.4

-0.2

0.0

0.2H

eat

Flo

w (

W/g

)

0 50 100 150 200 250

Temperature (°C)Exo Up Universal V4.2E TA Instruments

Micronized 60 psi

Micronized 20 psi

Thermal Analysis Comparison of API with Varied Micronization Energy

Amorphous component (exotherm) increases with increased micronization energies

Gravimetric Vapor Sorption

Re-crystallization

Post Gravimetric Vapor Sorption

Post GVS

After one GVS cycle the re-crystallization event is gone –consistent with amorphous recrystallization

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

Hea

t F

low

(W

/g)

20 40 60 80 100 120 140 160 180 200

Temperature (°C)

034 MICRONIZED dry 20psi rerun 20Feb.001034 MICRONIZED DRY 20PSI Post GVS RERUN 20FEB.002

Exo Up Universal V4.2E TA Instruments

Separating Components of Disorder

Amorphous

Thermal analysisGVS-recrystallization

SS-NMR

Nano-crystalline content

XRPD – line broadeningGVS/porositySurface area

Why XRPD Line Broadening?

Why study micronized materials by XRPD?

Crystal is essentially a 3D molecular diffraction grating

http://physics-animations.com/Physics/English/DG10/

The Scherrer Equation

cos

2L

KB

Peak width (B) is inversely proportional to the crystallite size (L)

P. Scherrer, “Bestimmung der Grösse und der inneren Struktur von Kolloidteilchen mittels Röntgenstrahlen,” Nachr. Ges. Wiss. Göttingen 26 (1918) pp 98-100.

Short Range vs. Long Range Order

XRPD – Particle vs. Crystallite

XRPD – Instrumental Factors

Same sample – Different instrument settingsNeed quality and consistent data collection

capillary

Capillary sample may be loaded into holder (pre-aligned) and analyzed in standard transmission setup with auto-changer and without need for instrument

reconfiguration to capillary analysis with standard goniometer heads.

Convenient Capillary Analysis Setup

Possible Spinning in both X-Y and X-Z

planes

XRPD – Sample Analysis

XRPD – Sample Analysis – Transmission Setup

Visual Comparison of Raw Patterns

5 10 15 20 25 30 35 40 45 50 552Theta (°)

0

20000

40000

60000

80000Inte

nsity

(cou

nts)

Profile Fitting To Obtain FWHM

Position [°2Theta]

10 20 30 40 50

Counts

0

20000

40000

x2813a1-RoomTempx2813a1 100.0 %

Position [°2Theta]

10 20 30 40 50

Counts

0

20000

40000

x2813a1-RoomTempx2813a1

10 20 30 40 50

Counts

0

20000

40000

FWHM

XRPD – Sample Analysis

Non-micronized and freshly micronized materials show significantdifferences in linewidth – consistent and suitable for analysis

“Annealing” Studies

Study 1 Samples of fresh micronized material were equilibrated at 6 different temperatures. Samples were heated in the DSC cell, in open DSC pans, and held isothermal for at least 60 min.

Annealing1. To subject (glass or metal) to a process of heating and slow cooling in order to toughen and reduce brittleness.2. To strengthen or harden.

Study 1 Results

Linear relationship observed with increasing temperature and XRPD linewidth (FWHM) – Increase in crystallite size as increase temperature

LiCl (~10 %) NaBr ( ~50 %) K2SO4 (~95%)Ambient Ambient Ambient

50 °C 50 °C 50 °C

Study 2 Samples of micronized material lot stored at 3 different RHs and at 2 different temperatures. Samples were pulled periodically over a several week period of time.

Environmental “Annealing”

Study 2 – Time Course

Within 2 weeks all conditions appear to have reached an equilibrium with respect to a reduced FWHM

Study 2 – Temperature Humidity Effect

A synergistic effect on annealing is apparent with increased temperature and relative humidity

Freshly Micronized “Annealed”

Microscope Comparison

No obvious difference in particles post “annealing” - Consistent with healing of fractures/fines.

Healing of Fractures

Increase in averaged crystallite size

“Fresh” “Aged”

“Fresh” “Aged”

Conclusions

With further study we may be able to better understand “disorder” generated in micronized API’s and separate the contributions of true glassy amorphous and nano-crystalline content.

We are currently evaluating the physiochemical changes that are associated with this observed crystal “healing” - related moisture sorption, surface area and porosity, thermal, XRPD, and ssNMR data.

We are applying this same methodology to other projects with similar observations/results.

Acknowledgements

Jeff Brum

Rachel Forcino

Fred Vogt (SS-NMR)

Robert Carlton (SEM)